Atherosclerotic aortic lesions increase the risk of cerebral embolism during carotid stenting in patients with complex aortic arch anatomy
Article Outline
Background
Carotid artery stenting (CAS) leads to frequent embolic brain lesions; their source has not been clearly identified yet. In order to investigate this phenomenon, we have evaluated embolic brain lesions (BL) after CAS and correlated them with aortic arch (AA) characteristics.
Methods
The AAs of 59 patients undergoing CAS under distal protection were evaluated by angiography and transesophageal echocardiography (TEE). AAs were stratified according to morphology (type I and II “simple” vs type III and bovine “difficult”), atherosclerotic arch lesions (complicated: >5 mm or with mobile debris vs uncomplicated: <5 mm), and tortuosity index (TI; sum of all angles diverging from ideal carotid axis, <150 vs >150). Diffusion weighted imaging (DWI) was performed before and within 24 hours from CAS. New BL were considered ipsilateral (IL) if ipsilateral to the site of CAS and non-ipsilateral (CL) if contralateral to it or bilateral. Normality distribution was by Shapiro-Wilk test (variables reported as medians ± interquartile range) and statistical significance (P < .05) by Wilcoxon and Fisher's exact test.
Results
Difficult arches were present in 17 patients (28.8%), complicated aortic plaque in 21 (35.5%), and TI > 150 in 34 (57.6%). New BL appeared in 34 or 57.6% patients (6 or 18% IL and 28 or 82% CL). The mean number of BL was 5.7 (range, 0 to 20), 4.7 IL, and 5.7 CL, with a median volume of 560.95 ± 1677.7 mm3. Type of arch and TI were not correlated with mean number of BL. Mean volume of BL were greater in patients with difficult AA, complicated plaques, and TI > 150 (258 (572) mm3 vs 15.6 (353) mm3, P = .2; and 86 (828) mm3 vs 85.9 (352) mm3, P = .4 172 (766) mm3 vs 0 (228) mm3, P = .06, respectively). In patients with all three AA characteristics, mean number and volume of BL was significantly greater compared with other patients. Specifically, this increase was due mainly to CL (IL 0 (117) mm3 vs 0 (172) mm3, P = .9; CL 564 (687) mm3 vs 0 (133) mm3, P = .001). None of the technical details considered was correlated with either IL or CL.
Conclusion
BL are frequent after protected CAS and are correlated with AA characteristics, thus underlining the role of catheterization maneuvers in determining embolic events. TEE may be useful in patient's selection for CAS.
Cerebral embolic damage occurs frequently during and after carotid artery stenting (CAS) procedures.1, 2, 3, 4, 5, 6, 7 Diffusion weight magnetic resonance imaging (DW-MRI) reveals presence of new embolic lesions in up to 70% of patients.8 Thus, cerebral embolization appears to be much more common after CAS than after carotid endarterectomy (CEA).8, 9
The precise cause of the embolism remains unclear. Some authors hypothesize that the embolism occurs during catheterization maneuvers or stenting deployment. However, the finding of late lesions has led other investigators to believe that emboli migrate from the carotid bifurcation after stent deployment. The observation that many of the lesions are contralateral to the site of CAS suggests the possibility of other factors.4, 5, 7, 8
In the present study, we used DW-MRI and transesophageal echocardiography (TEE) to explore possible associations between atherosclerotic aortic lesions, considered in the context of the variable anatomy of the aortic arch and its branches, and appearance of new cerebral embolic lesions after CAS.
Methods
Since 2006, all asymptomatic patients routinely undergoing CAS who gave informed consent received a pretreatment TEE, along with pre- and post-treatment cerebral DW-MRI.
TEE
TEE aortic examinations were conducted one to three days before CAS using a Hewlett-Packard Sonos 5500 sonograph (Hewlett-Packard, Philips, NJ) with a multiplane transesophageal probe (patients received oropharyngeal local anesthetic and mild sedation with intravenous midazolam).
DW-MRI
DW-MRI scans were obtained both one to three days before CAS and during the first 24 hours after the procedure. We used a 1.5T General Electrics Medical Systems (Milwaukee, Wisc) Signa Horizon LX whole-body scanner with a quadrature birdcage headcoil. A fluid-attenuated inversion recovery (FLAIR) sequence was performed in the oblique-axial plane (repetition time, TR; 8002 ms; echo time; TE, 116 milliseconds, inversion time, TI 2000 seconds; field of view [FOV] 24 × 24 cm; thickness, 5 mm; inter-slice gap, 0 mm; matrix, 256 × 224). Axial DW images were obtained (thickness, 5 mm; inter-slice gap, 0 mm) using a single-shot Echo-planar-imaging (EPI) sequence (matrix size, 192 × 192; FOV, 32 × 32 cm; 3 next), as previously reported.10 In-plane spatial resolution was 2.77 mm2. Diffusion encoding gradients were applied in three orthogonal directions, with gradient strength corresponding to b-values of 900 mm2/seconds. We additionally acquired images without diffusion weighting (corresponding to b = 0 mm2/seconds and exhibiting a T2-contrast).
Pre-CAS DW-MRI images were automatically mapped onto the post-CAS images by rigid body registration (flexible image registration toolbox [FLIRT], fMRIB software library [FSL], Oxford center for functional MRI of the brain [fMRIB], Oxford11). Hyperintense lesions were visually identified on the post-CAS DW-MRI images. Identification was confirmed by the absence of hyperintensity on the corresponding pre-CAS DW-MRI images. The volume and the number of lesions were assessed semi-automatically, with final revision by one of the specialized physicians involved in the study (R.L., C.T., B.B.). The volume of the lesions was quantified by defining a threshold for the difference between pre-CAS and post-CAS image intensities, and counting the number of pixels where the difference was greater than this threshold.
Classification of new brain lesions was based on the vascular territory in which they occurred. New lesions were defined “ipsilateral (IL)” if located in the hemisphere ipsilateral to the side of the treated carotid bifurcation; or “non-ipsilateral (CL)” if located in the hemisphere contralateral to it, in the posterior territory, or in both the ipsilateral and contralateral hemisphere.
CAS
CAS was conducted as previously described12, 13 by an experienced vascular surgeon with a case load of more than 300 CAS performed (G.L.F.). Briefly, patients were taken to the angiographic suite following appropriate informed consent and cardiological evaluation and medicated with aspirin 100 mg and clopidogrel 75 mg for 3 days before the procedure. All procedures were performed under local anesthesia, systemic heparinization, 8F groin introducer. Common carotid cannulation was achieved with 40 degree Boston Scientific (Natick, Mass) or Medtronic (Minneapolis, Minn) HS I and II catheters over Terumo (Tokyo, Japan) stiff guidewire. When cannulation was not achievable by these means, several different alternate techniques were used (ie, buddy wire, coaxial). Brachial or carotid access was not attempted in any case. Routine cerebral protection was by Filterwire EZ (Boston Scientific), Angioguard RX (Cordis, Warren, NJ) or Accunet RX (Guidant, St Paul, Minn) and stenting by Wallstent (Boston Scientific), Precise (Cordis) or Acculink (Guidant). “Technical success” was defined as the capacity of treating the stenosis with residual stenosis of less than 30%. Neurological outcome was evaluated both at the end of the procedure and in the next 24 hours by a neurologist according to National Institutes of Health stroke scale and the modified Rankin scale. Neurological, general, and technical (duplex imaging) outcome was evaluated at discharge and at 30 days, 6 months, and yearly thereafter.
Study definitions and data analysis
The main outcome measures were the number and volume of new brain lesions at DW-MRI, with the volume being the most accurate measurement method in ischemic stroke evaluation.14 Risk factors for new brain lesions were studied in terms of aortic arch anatomy, atherosclerosis of the aortic arch, and tortuosity of the supraortic vessels. Aortic arches were categorized into “simple arches” (type I and type II)15 and “difficult arches” (type III or bovine arches). Atherosclerotic plaques of the aortic arch seen at TEE were categorized as “complicated” in the presence of mobile debris or an endoluminal protrusion > 0.5 mm; other plaques were considered “uncomplicated.” Tortuosity of the supraortic vessels was assessed by the “tortuosity index” (TI), calculated as the sum of all angles diverging from the ideal straight axis, as described elsewhere.12 Based on the combination of these three variables, we also divided patients in those with all three arch characteristics (difficult arch, complicated aortic plaques, and TI > 150) vs those without all three characteristics.
Normality distribution was assessed using the Shapiro-Wilk test, and all variables that were non-normally distributed were reported in terms of medians ± interquartile range (IQR). For the analysis, all patients without cerebral lesions have been estimated at the value of zero. Differences between groups were evaluated using Fisher's exact test for categorical variables and the Wilcoxon rank-sum test for continuous ones. Statistical significance was defined as a two-sided P
value < .05, and all analyses were carried out using STATA statistical software, version 8.2 (Stata Corporation, College Station, Tex) by an expert statistician (L.M.).
Results
Table I summarizes the characteristics of the 59 patients who were eligible for analysis. All patients in this series were asymptomatic for cerebral ischemia; preoperative DW-RMI was negative in all cases. Difficult arch was present in 17 (29%) patients, TI >150 in 34 (58%), and complicated aortic plaques in 21 (36%).
Table I. Epidemiology of the study sample according to aortic arch characteristics
| Mean age (SD) years | Male gender n (%) | Hypertension n (%) | Current smoking n (%) | Chronic obstructive pulmonary disease n (%) | Diabetes mellitus n (%) | Chronic renal failure (GFR<60 ml/min) n (%) | Dislipydemia n (%) | Coronary artery disease n (%) | |
|---|---|---|---|---|---|---|---|---|---|
| Total sample (n = 59) | 75.86(4.12) | 33(56) | 49(83) | 7(12) | 13(22) | 13(22) | 10(17) | 18(30) | 21(35) |
| Simple arch (n = 42) | 75.64(3.8) | 29(69) | 36(86) | 6(14) | 8(19) | 8(19) | 6(14) | 13(31) | 16(38) |
| Difficult arch (n = 17) | 76.41(4.7) | 4(24) | 13(77) | 1(6) | 5(29) | 5(29) | 4(24) | 5(29) | 5(29) |
| Uncomplicated aortic plaque (n = 38) | 76.13(4.5) | 24(63) | 30(79) | 6(16) | 5(13) | 7(18) | 6(16) | 10(26) | 10(26) |
| Complicated aortic plaque (n = 21) | 75.3(3.1) | 9(43) | 19(90) | 1(5) | 8(38) | 6(29) | 4(19) | 8(38) | 11(52) |
| TI <150 (n = 25) | 75.6(4.8) | 16(64) | 17(68) | 3(12) | 6(24) | 5(20) | 3(12) | 2(8) | 6(24) |
| TI >150 (n = 34) | 76(3.5) | 17(50) | 32(94) | 4(12) | 7(21) | 8(24) | 7(21) | 16(47) | 15(44) |
| No complex arch (n = 53) | 75.9(4.2) | 33(62) | 43(81) | 7(13) | 10(19) | 10(19) | 8(15) | 15(28) | 17(32) |
| Complex arch (n = 6) | 75.3(4.0) | 6(100) | 6(100) | 0(0) | 3(50) | 3(50) | 2(33) | 3(50) | 4(67) |
No case of procedure-related death or stroke occurred. Two patients experienced a transitory ischemic attack in the ipsilateral hemisphere (1 and 5 hours after the procedure, respectively), as defined by neurological examination.
Thirty-four (58%) patients showed a total of 180 new lesions at DW-MRI. (IL in 6 patients or 18%, and CL in 28 or 82%). Mean volume of the lesion was 560.95 ± 1677.7 mm3. In the entire study population, no association was seen between the mean number of new lesions and tortuosity and type of arch (Table II). The presence of complicated aortic plaques determined a greater number of inconisistent lesions (Table II). Mean lesion volume was also independent to these variables; however, there was a trend toward greater volume in groups with type III or bovine arches and complicated aortic arches (Table III).
Table II. Mean lesion number at diffusion weight magnetic resonance imaging after carotid artery stenting according to aortic arch characteristics
| Difficult arch (n = 17) | Simple arch (n = 42) | P⁎ | |
|---|---|---|---|
| Total lesion number | 2(6) | 1(2) | .2 |
| Ipsilateral lesion number | 6(0)⁎⁎ | 1(1) | .11 |
| Non-ipsilateral lesion number | 2(7) | 2(3) | .2 |
| Complicated aortic plaque (n=21) | Uncomplicated aortic plaque (n=38) | ||
| Total lesion number | 2(8) | 1(2) | .087 |
| Ipsilateral lesion number | 1(0)⁎⁎ | 2(1) | .3 |
| Non-ipsilateral lesion number | 4(5) | 1(1) | .047 |
| TI >150 (n=34) | TI < 150 (n=25) | ||
| Total lesion number | 1(6) | 0(2) | .096 |
| Ipsilateral lesion number | 2(1) | 0(0)⁎⁎⁎ | — |
| Non-ipsilateral lesion number | 2(5) | 2(1) | .4 |
| All three characteristics together§(n=6) | Other patients (n=53) | ||
| Total lesion number | 7(7) | 1(2) | .01 |
| Ipsilateral lesion number | 0(0)⁎⁎⁎ | 1(1) | — |
| Non-ipsilateral lesion number | 7(8) | 2(1) | .051 |
⁎Normality distribution was assessed using Shapiro-Wilk test. All variables were non-normally distributed and were thus reported in terms of medians ± Interquartile range (IQR) in parenthesis. Significance was assessed using Wilcoxon rank-sum test for independent observations. |
⁎⁎Only one of these patients had consistent lesions. |
⁎⁎⁎None of these patients had this type of lesions. |
§Patients with difficult arch, complicated aortic plaque, and tortuosity index (TI) > 150. |
Table III. Volumes of cerebral lesions at diffusion weight magnetic resonance imaging after carotid artery stenting according to aortic arch characteristics
| Difficult arch (n = 17) | Simple arch (n = 42) | P⁎ | |
|---|---|---|---|
| Total volume (mm3) | 258(572) | 15.6(353) | .2 |
| Volume, ipsilateral lesions (mm3) | 0(117) | 0(172) | .5 |
| Volume, non-ipsilateral lesions (mm3) | 133(365) | 0(93.7) | .027 |
| Complicated aortic plaque (n=21) | Uncomplicated aortic plaque (n=38) | ||
| Total volume (mm3) | 86.0(828) | 85.9(352) | .4 |
| Volume, ipsilateral lesions (mm3) | 0(148) | 0(172) | .8 |
| Volume, non-ipsilateral lesions (mm3) | 78.1(400) | 15.6(93.7) | .2 |
| TI>150(n=34) | TI<150(n=25) | ||
| Total volume(mm3) | 172(766) | 0(228) | .063 |
| Volume, ipsilateral lesions (mm3) | 0(278) | 0(0) | .087 |
| Volume, non-ipsilateral lesions (mm3) | 82.0(312) | 0(180) | .19 |
| All three characteristics together§(n=6) | Other patients (n=53) | ||
| Total volume (mm3) | 564(570) | 31.2(352) | .018 |
| Volume, ipsilateral lesions (mm3) | 0(117) | 0(172) | .9 |
| Volume, non-ipsilateral lesions (mm3) | 564(687) | 0(133) | .001 |
⁎Normality distribution was assessed using Shapiro-Wilk test. All variables were non-normally distributed and were thus reported in terms of medians ± Interquartile range (IQR). Significance was assessed using Wilcoxon rank-sum test for independent observations. |
§Patients with difficult arch, complicated aortic plaque and tortuosity index (TI) > 150. |
By considering patients with all three characteristics (“complex arch”) (ie, TI > 150, type III or bovine arch, and complicated plaques), both the mean lesion number and mean lesion volume were significantly greater compared with other patients. This difference was mainly due to CL (Table II, Table III).
No correlations were found between the number, size, and volume of new lesions on DWI and other variables (age, gender, carotid plaque morphology, grade of carotid stenosis, type of stent and protection device, and pre-existing cerebral lesions, type of protection device, and stent). The two patients with postoperative TIA had type I arch in one case and a bovine arch in the other and showed CL at DWI in both cases.
Discussion
While confirming the high frequency of new cerebral lesions on DW-MRI following clinically uneventful CAS procedure, this study suggests the hypothesis of a pathogenetic association between the degree of atherosclerotic changes of the aortic arch (as evaluated at TEE) and the risk of periprocedural embolism, at least in patients with complex aortic arch anatomy.
In previous studies using systematic pre- and post-CAS cerebral DW-MRI, “new lesions,” both at ipsilateral and contralateral level, were apparent in 15% to 70% of the patients (Table IV). In particular, the frequency of patients with new ischemic lesions in the hemisphere contralateral to the CAS site or in the posterior circulation (CL) ranged from 0% to 52% (Table IV). The higher values recorded in the present work can be attributed to the sensitivity of the DW-MRI protocol used and to the relatively high number of patients with a “complex” aortic arch (ie, challenging/severe atherosclerotic lesions in the context of a difficult aortic arch).
Table IV. Patients with cerebral lesions after carotid artery stenting at diffusion weighted imaging from the literature
| Author | N cases studied, total | N cases with new lesions | Consistent | Inconsistent |
|---|---|---|---|---|
| van Heesewijk, 20022 | 72 | 11(15%) | 15% | 0% |
| Flach, 20049 | 21 | 9(43%) | — | — |
| Hammer, 20054 | 53 | 21(40%) | 15.1% | 24.5% |
| Roh, 200517 | 22 | 8(36%) | — | — |
| Hauth, 20051 | 105 | 22(21%) | — | — |
| de Rochemont, 20063 | 50 | 14(28%) | — | — |
| Asakura, 200618 | 45 | 20(44%) | — | — |
| Rapp, 20075 | 54 | 36(67%) | 97% | 28% |
| Tedesco, 20078 | 27 | 19(70%) | 47% | 52% |
| El-Koussy, 20076 | 44 | 13(30%) | — | — |
| Barbato, 20087 | 35 | 21(60%) | 53% | 19% |
| Present study | 59 | 34(57.6%) | 18% | 82% |
Appearance of ipsilateral embolisms can be at least partially attributed to detachment of particles from the carotid plaque during stenting, providing the rationale for the use of distal filters. However, the source of contralateral lesions (CL) and the pathogenetic mechanism underlying their formation remain to be elucidated. Indirect evidence that the aortic arch can be a source of embolization comes from a study of Barbato et al,7 where occurrence of cerebral microemboli during CAS was not prevented by the use of a filter distal to the stented carotid artery, and in a fifth of the cases the microemboli occurred in the contralateral hemisphere. Thoracic computed tomography (CT) scan findings have shown statistical relations between advancing age, increasing aortic arch calcium content and presence of a type II arch.16 These findings led Bazan et al to suggest that high aortic arch calcium content and type II arches may be markers of increased potential for embolization during arch manipulation.16 A previous study from our group indicated that both aortic arch anatomy and the presence of tortuosity of the carotid artery proximal to the lesion significantly increase the risk of both technical failure and neurological complications during CAS.12, 13 From the limited information available in the literature, it is reasonable to hypothesize that “complicated” (ie, protruding, mobile) atherosclerotic lesions of the aortic arch can be an important source of CAS-related cerebral embolism. To explore this concept, we decided to use TEE to detect and quantify atherosclerotic lesions in the aortic arch since this technique allows precise evaluation of the extent and composition of aortic plaques while avoiding the limitations of contrast CT (including nephrotoxicity, overestimation of calcific plaques, and imprecise sizing of plaque thickness). To our knowledge, this is the first time TEE evaluation has been used in this investigative context. While no significant correlation between severity of aortic atherosclerosis and amount of cerebral embolization was apparent in the overall study population, among patients with “complex” aortic arch anatomy and more tortuous carotid arteries presence of complicated plaques (> 5 mm or with mobile debris) was associated with the number and volume of cerebral lesions. Remarkably, this association was largely driven by the appearance of new lesions in the hemisphere contralateral to the target carotid stenosis or in the posterior circulation territory. This observation seems to support the possible pathogenetic role of catheter and wire manipulation within the arch.
Although in our practice both symptomatic and asymptomatic patients are treated by CAS,12, 13 this study was intentionally limited to asymptomatic patients in order to avoid possible confounding overlapping with pre-existing lesions. Our analysis was limited by the small number of patients presenting all the three characteristics of arch complication (ie, difficult arch, complicated plaque, tortuosity index >150). However, the strong statistical significance of the outcome in this setting, together with the observation that these patients displayed high number of contralateral or posterior cerebral lesions, points to the relevance of a field where conclusion on the pathogenetic mechanism of complications are yet to be drawn.
Our study was not designed to analyze the timing of the occurrence of the cerebral lesions. Some authors suggest that embolic events may occur at any time during and after CAS.5 The two symptomatic neurological events in ours series (TIA in the hemisphere ipsilateral to the stented carotid lesion) manifested a few hours after the procedure. A plausible explanation for late embolization phenomena (after completion of CAS) is that fragments which are only partially dislodged from the arch during the procedure may subsequently detach and migrate to the brain. Obviously, late embolization from the plaque in the bifurcation through the stent cells remains a possibility.
Conclusion
Our study suggests that in patients with complex aortic arch anatomy and carotid tortuosity, in whom catheterization maneuvers are likely to be more prolonged, there may be a pathogenetic association between the possible presence of protruding or mobile atherosclerotic plaques of the aortic arch and the risk of periprocedural embolization. Thus, although TEE may not be necessary on a routine basis, it might have a role in estimating clinically relevant risks and benefits associated with CAS in patients with complex arch anatomy and carotid tortuosity. The presence of complicated plaque in these patients should therefore contraindicate CAS through an aortic arch route.
Author contributions
References
- MR and clinical follow-up of diffusion-weighted cerebral lesions after carotid artery stenting. AJNR Am J Neuroradiol. 2005;26:2336–2341
- New brain lesions at MR imaging after carotid angioplasty and stent placement. Radiology. 2002;224:361–365
- . Diffusion-weighted MR imaging lesions after filter-protected stenting of high-grade symptomatic carotid artery stenoses. AJNR Am J Neuroradiol. 2006;27:1321–1325
- Cerebral microembolization after protected carotid artery stenting in surgical high-risk patients: results of a 2-year prospective study. J Vasc Surg. 2005;42:847–853
- Subclinical embolization after carotid artery stenting: new lesions on diffusion-weighted magnetic resonance imaging occur postprocedure. J Vasc Surg. 2007;45:867–872
- Periprocedural embolic events related to carotid artery stenting detected by diffusion-weighted MRI: comparison between proximal and distal embolus protection devices. Endovasc Ther. 2007;14:293–303
- A randomized trial of carotid artery stenting with and without cerebral protection. J Vasc Surg. 2008;47:760–765
- Postprocedural microembolic events following carotid surgery and carotid angioplasty and stenting. J Vasc Surg. 2007;46:244–250
- Cerebral ischemia after carotid intervention. J Endovasc Ther. 2004;11:251–257
- Diffusion MRI shows increased water apparent diffusion coefficient in the brains of cirrhotics. Neurology. 2004;62:762–766
- . A global optimisation method for robust affine registration of brain images. Med Image Anal. 2001;5:143–156
- Measurement and impact of proximal and distal tortuosity in carotid stenting procedures. J Vasc Surg. 2007;46:1119–1124
- Aortic arch anomalies are associated with increased risk of neurological events in carotid stent procedures. Eur J Vasc Endovasc Surg. 2007;33:436–441
- . Acute ischemic stroke lesion measurement on diffusion weighted imaging – Important considerations in designing acute stroke trials with magnetic resonance imaging. J Stroke Cerebrovas Dis. 2007;16:64–70
- . Regional anatomy of the thoracic aorta. In: Kadir S editors. Atlas of normal and variant angiographic anatomy. Philadelphia: WB Saunders; 1991;p. 19
- . Increased aortic arch calcification in patients older than 75 years (Implications for carotid artery stenting in elderly patients). J Vasc Surg. 2007;46:841–845
- Prospective analysis of cerebral infarction after carotid artery stent placement by using diffusion-weighted imaging. AJNR Am J Neuroradiol. 2005;26:376–384
- Diffusion-weighted magnetic resonance imaging in carotid angioplasty and stenting with balloon embolic protection devices. Neuroradiology. 2006;48:100–112
Competition of interest: none.
PII: S0741-5214(08)01364-5
doi:10.1016/j.jvs.2008.08.014
© 2009 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
