Volume 39, Issue 3 , Pages 252-257, March 2010
Cervical Access for Filter-protected Carotid Artery Stenting: A Useful Tool to Reduce Cerebral Embolisation
Article Outline
Abstract
Background
Filter-protected transcervical carotid artery stenting (CAS) has been suggested to reduce the intraoperative cerebral embolisation observed during transfemoral CAS. We therefore evaluated clinical outcome and incidence of ischaemic lesions at diffusion-weighted magnetic resonance imaging (DW-MRI) after transcervical and transfemoral CAS.
Methods
From March 2007 to May 2009, we performed filter-protected CAS in 135 patients with symptomatic (30%) or asymptomatic (70%) carotid stenosis above 70% and below 95%. In 44 patients with risky femoral access or unfavourable aortic arch anatomy, access to common carotid artery was achieved by a small cervical incision. In another 91 procedures we used a classic percutaneous femoral access. Preoperative and postoperative DW-MRI scans were obtained after 111 procedures (82%) – 35 transcervical and 76 transfemoral.
Results
The incidence of clinical events (transient ischaemic attack (TIA) and stroke) was 2.3% after transcervical CAS and 19.8% after transfemoral CAS (P
<0.01), without any deaths. DW-MRI disclosed new ischaemic lesions in five patients (5/35, 14.3%) after transcervical CAS and in 28 patients (28/76, 36.8%) after transfemoral CAS (P=0.015). All ischaemic lesions depicted after transcervical procedures were ipsilateral to the treated artery.
Conclusions
Transcervical filter-protected CAS, compared with classic percutaneous procedures, seems to reduce clinical events and DW-MRI ischaemic damage and may be useful in selected patients.
Keywords: Carotid arteries, Carotid stenosis, Stents, Aortic arch, Diffusion-weighted MRI
Introduction
Transfemoral carotid artery stenting (CAS) has emerged in these past years as a possible alternative to carotid endarterectomy for treating patients with carotid stenosis.1 Cerebral protection systems, mainly distal filters, are currently used and seem able to reduce stroke and death rates to values almost overlapping those observed after surgery.2, 3
Silent cerebral embolisation during CAS nevertheless remains an open problem insofar as numerous studies using diffusion-weighted magnetic resonance imaging (DW-MRI) document a high incidence of new asymptomatic postprocedural cerebral ischaemic lesions.4, 5 The report of ischaemic lesions in cerebral territories not directly involved by the carotid procedure6 or after a merely diagnostic angiographic study of the supra-aortic trunks7 suggests that a non-negligible part of the procedural embolic load could be linked to aortic arch instrumentation inherent in transfemoral procedures. A higher incidence of adverse clinical events8 and postoperative DW-MRI ischaemic lesions9 was indeed reported in octogenarians, known to have increased prevalence of aortic arch atherosclerosis and complex anatomy.10 Moreover, as transfemoral access for CAS may not be feasible in patients with severe and bilateral aorto-iliac occlusive disease, common carotid cannulation may also be impossible in patients with tortuous supra-aortic trunks or unfavourable aortic arch. In a recent article evaluating the impact of tortuosities on transfemoral CAS, a technical failure related to aortic arch or common carotid artery tortuosities was reported in 9% of the procedures, showing that this eventuality should not be underestimated.11
To reduce the inherent embolic load and to overcome possible difficulties linked to femoral access, some have proposed to perform filter-protected CAS through a cervical access, either percutaneous or surgical.12, 13, 14 Even if reported clinical outcomes were satisfactory, no DW-MRI data or comparison with transfemoral filter-protected CAS was provided. Hence, we undertook this non-randomised observational study to compare transcervical and transfemoral filter-protected CAS in terms of clinical outcome and procedural embolisation rate, evaluated by DW-MRI.
Materials and methods
From March 2007 to May 2009, we performed filter-protected carotid stenting in 135 patients. In 44 of these patients the procedure was done through a small cervical cut-down while in the other 91 patients we used a classic percutaneous transfemoral approach. Baseline characteristics of patients included in this study are summarised in Table 1. The transcervical approach was always preferred in cases of difficult or risky femoral approaches (severe and bilateral aorto-iliac occlusive disease, large abdominal aortic aneurysm (>5cm) or presence of an aortobifemoral prosthesis). In presence of a complex anatomy of aortic arch or supra-aortic trunks (type III arch, bovine arch, common carotid artery (CCA) coiling or kinking), a cervical access was chosen according to the operating surgeon's experience in endovascular procedures: while the vascular surgeon with a broad experience always chose the femoral route, another two vascular surgeons at the beginning of their experience in carotid stenting preferred cervical access. Femoral access and anatomic features of patients undergoing transcervical and transfemoral CAS are listed in Table 2. Contraindications to cervical access were circumferential and diffused atheromatous disease of the CCA or an anatomically low carotid bifurcation (less than 5cm above the clavicle). No patients in these series were excluded for these reasons. Carotid stenting was indicated in presence of a symptomatic or asymptomatic stenosis of the internal carotid artery (ICA)≥70%. To avoid a risky crossing or an unprotected predilation of the lesion, patients with endoluminal thrombus or tight stenosis (>95%) were excluded. Another exclusion criteria was the presence of an excessive tortuosity of the distal ICA precluding proper filter deployment. Patients excluded from stenting were referred to surgery. Written informed consent was obtained from all patients, and the procedures were approved by the hospital ethics committee.
Table 1. Baseline characteristics of the 135 patients included in the study.
| Characteristic | Transcervical CAS | Transfemoral CAS | P |
|---|---|---|---|
| N° patients | 44 | 91 | |
| Mean age in years (range) | 72.1 (57–82) | 74.1 (59–88) | NS |
| Age ≥80years (%) | 7 (15.9) | 17 (18.7) | NS |
| Sex | |||
| Men (%) | 33 (75) | 59 (65) | NS |
| Women (%) | 11 (25) | 32 (35) | NS |
| Carotid stenosis | |||
| 70–80% (%) | 24 (55) | 35 (38) | NS |
| 81–95% (%) | 20 (45) | 56 (62) | NS |
| Type of carotid plaque | |||
| Type 1–2 (%) | 11 (25) | 23 (25) | NS |
| Type 3–5 (%) | 33 (75) | 68 (75) | NS |
| Contralateral carotid stenosis | |||
| <50% (%) | 26 (59) | 61 (67) | NS |
| 50–99% (%) | 16 (36) | 27 (30) | NS |
| Occlusion (%) | 2 (5) | 3 (3) | NS |
| Clinical presentation | |||
| Symptomatic (%) | 14 (32) | 27 (30) | NS |
| Asymptomatic (%) | 30 (68) | 64 (70) | NS |
Table 2. Femoral access and anatomic issues in patients undergoing transcervical CAS and transfemoral CAS.
| Characteristics | N° patients undergoing transcervical CAS (%) | N° patients undergoing transfemoral CAS (%) | P |
|---|---|---|---|
| Femoral access issues | |||
| Severe bilateral aorto-iliac occlusive disease | 3 (8) | 0 | 0.03 |
| Abdominal aortic aneurysm (>5cm) | 3 (8) | 0 | 0.03 |
| Aortobifemoral prosthesis | 2 (5) | 0 | 0.1 |
| Anatomic issues | |||
| Type III aortic arch | 8 (18) | 6 (7) | 0.06 |
| Bovine aortic arch | 7 (16) | 5 (5.5) | 0.06 |
| Common carotid artery tortuosity | 11 (25) | 11 (12) | 0.08 |
| Aortic arch calcifications | 10 (23) | 9 (10) | 0.06 |
| No significant issues | 0 | 60 | <0.01 |
| Total | 44 | 91 | |
Except for one patient who had a restenosis from a previous endarterectomy, all patients had primary atherosclerotic lesions of the ICA, as documented on ultrasound. The degree of carotid stenosis was evaluated according to velocimetric and morphologic criteria. Plaque structure was assessed by B-mode ultrasonography and defined according to the Geroulakos15 criteria: type 1: uniformly echolucent; type 2: predominantly echolucent; type 3: predominantly echogenic; type 4: uniformly echogenic; and type 5: not classifiable owing to heavy calcification producing acoustic shadow. All patients underwent preoperative magnetic resonance angiography (MRA) or computed tomographic angiography (CTA) to study the aortic arch and the supra-aortic vessels.
All patients were taking acetylsalicylic acid before the intervention and had an oral loading dose of clopidogrel (300mg) at least 3h before the procedure. An intravenous (IV) bolus of heparin (80IU kg−1) was given after sheath placement. During the procedure, if an activated clotting time of less than 250s was found, the patient received an additional dose of heparin.
Prior to all transcervical procedures, the skin was marked at the level of the carotid bifurcation under ultrasound guidance. Then, in the operating room, a needle was transversally positioned over the skin mark to identify the carotid bifurcation under fluoroscopic guidance. In all patients, under local anaesthesia, a small longitudinal incision, with an estimated length of 2cm, was made along the sternocleidomastoid muscle above the clavicle. The CCA was dissected circumferentially and encircled with a loop. Then a 6 F×7-cm long introducer sheath (‘Super Sheath’, Togo Medikit Co., Tokyo, Japan) was inserted under fluoroscopy over a 0.035-inch guidewire into the CCA, taking care not to go beyond the bifurcation needle. Because of its short length, no continuous pressurised heparinisation of the sheath was deemed necessary. After confirming ICA stenosis with an angiogram, a cerebral protection filter (‘Filterwire EZ’, Boston Scientific Corporation, Mountain View, USA) was placed within the distal ICA. Predilation of the stenosis with a coronary-artery balloon was necessary in three patients. Then, a self-expanding stent (‘Carotid Wallstent Monorail’, Boston Scientific Corporation, Natick, MA, USA; ‘Precise Carotid Stent’, Cordis Corporation, Miami, FL, USA) was released at the carotid bifurcation and post-dilated with a 5- or 6-mm diameter balloon (‘Ultrasoft SV’, Boston Scientific Corporation, Natick, MA, USA). Prophylactic intravenous atropine (1mg) was routinely administrated before balloon dilation. A completion angiogram was obtained to confirm satisfactory treatment of the stenosis and the absence of spasm in the distal ICA. Intra-arterial nitroglycerin (200μg) was injected when a spasm occurred. A 6/0 Prolene stitch was then placed at the arterial puncture site to achieve haemostasis.
All percutaneous femoral procedures were performed under local anaesthesia. Common carotid artery cannulation was achieved through a telescopic technique with a 5 or 6 F 90-cm long sheath (‘Flexor Check-Flo Introducer’, Cook, Bloomington, IN, USA), a 5 F diagnostic catheter with an appropriate shape and a 0.035-inch standard hydrophilic guidewire (‘Zipwire’, Boston Scientific Corporation, Natick, MA, USA). Once the long sheath was placed into the common carotid artery, a pressurised bag of heparinised saline was connected to its proximal end to prevent thrombosis. Filter-protected carotid stenting was then performed as previously described for transcervical procedures. Predilation of the stenosis with a coronary-artery balloon was necessary in six patients. After femoral sheath withdrawal, haemostasis was achieved by manual compression.
MRI sequences were acquired at 1.5T with a dedicated brain imaging superconducting magnet. T2-weighted fluid-attenuation inversion recovery (FLAIR) sequences were acquired (repetition time (RT): 8000ms; echo time (ET): 100ms; inversion time: 2200ms; field-of-view: 240mm; turbo factor: 14; and number of excitations: 2) and DW sequences were acquired in the three planes at a thickness of 5mm (RT 3000, ET 84ms, b-value 0, 500 and 1000s mm−2). Patients underwent MRI within 3days before and after the procedure. When the MRI scanner was not available within 3days or MRI was contraindicated (patients with pacemakers or suffering claustrophobia), the patient was excluded from MRI. Two expert neuroradiologists, not involved in the CAS procedures, compared pre-procedural and postprocedural images and assessed the presence of recent ischaemic lesions. Ipsilateral lesions were defined as those in the operated carotid artery territory and contralateral lesions as those in the non-operated carotid artery and vertebrobasilar artery territories.
All patients underwent a neurologic examination before and after the procedure. New neurologic deficits lasting more than 24h were defined as a stroke. Qualitative data were analysed for statistical significance with chi-square test or Fisher's exact test as appropriate and quantitative data with Student's t-test. P values less than or equal to 0.05 were considered to indicate statistical significance.
Results
The procedure was completed in all cases after transcervical CAS. Two transfemoral procedures had to be converted to transcervical CAS because of the impossibility to cannulate the CCA; thus the rate of completed procedures was 97.8%. These patients were excluded from our study because it was not possible to assign postoperative ischaemic lesions to either of the procedures. We used a carotid Wallstent in 114 patients with a straight carotid artery course whereas the more flexible Precise nitinol stent was preferred in 21 patients with tortuous carotid bifurcations. In all cases intraoperative angiography showed a resolution of the stenosis (residual stenosis <20%).
During the immediate postoperative course (<30days), none of the patients in either group died. We observed five strokes, all occurring after transfemoral CAS and one involving the contralateral hemisphere (5/91, 5.5% in the transfemoral group vs. 0% in the transcervical group, P=0.17, by Fisher's exact test). In all these patients, an ischaemic lesion was found at postoperative CT scan (one patient) or DW-MR (four patients). We also observed 14 transient ischaemic attacks (TIAs), one in the transcervical group and 13 in the transfemoral group. All patients with TIA underwent postoperative DW-MR, but in five cases, this imaging technique did not show any ischaemic lesion. The difference between the overall rate of neurological events (TIA and stroke) in the two groups was statistically significant (18/91, 19.8% vs. 1/44, 2.3%; P<0.01, by the chi-square test). If we exclude the five patients who had a postoperative TIA after transfemoral CAS but no detectable lesion at DW-MR, the difference remains significant (13/91, 14.3% vs. 1/44, 2.3%; P=0.03, by Fisher's exact test). One patient, with previously asymptomatic coronary-artery disease, showed myocardial ischaemic signs after transfemoral CAS and underwent urgent myocardial revascularisation. No cranial nerve deficits, ICA dissections or cervical haematomas were observed in transcervical patients. Similarly, no groin haematomas developed after transfemoral procedures. A distal ICA spasm developed in five transfemoral and two transcervical CAS patients but responded in all cases to intra-arterial nitroglycerine.
Of the 135 patients who underwent filter-protected CAS, preoperative and postoperative DW-MRI of the brain was performed in 111 patients (82%) (Table 3). Of these 111 patients, 76 underwent transfemoral CAS (76/91, 83.5%) and 35 transcervical CAS (35/44, 79.5%). The comparison of preoperative and postoperative DW-MRI of the brain disclosed new ischaemic lesions in five patients who underwent transcervical CAS (5/35, 14.3%) and in 28 patients who underwent transfemoral CAS (28/76, 36.8%), with a statistically significant difference between the two groups (P=0.015 by the chi-square test). Among these patients with new DW-MRI ischaemic lesions, postoperative neurological symptoms were observed in one patient of the transcervical group and in 12 patients of the transfemoral group (1/5, 20% vs. 12/28, 43% P>0.05 by Fischer's exact test). In the 76 patients who underwent transfemoral CAS, 26 had an anatomical issue and 50 not. The incidence of postoperative DW-MRI lesions was significantly higher in the subgroup with complex anatomy (23/26, 88%) compared with the subgroup without (5/50, 10%) (P<0.01, by Fisher's exact test). In the transcervical group, DW-MRI showed 10 new ischaemic lesions, all ipsilateral. In the transfemoral group, DW-MRI detected 85 new ischaemic lesions, 66 (78%) ipsilateral and 19 (22%) contralateral (Table 3). Lesions were located in cortical territories (76 lesions, 80%), subcortical territories (13 lesions, 14%) or deep territories (six lesions, 6%). No ischaemic lesions were recorded after CAS done with predilation.
Table 3. Diffusion-weighted magnetic resonance results in patients undergoing transcervical and transfemoral carotid stenting.
| Transcervical CAS (n=35) | Transfemoral CAS (n=76) | P | |
|---|---|---|---|
| Patients with postoperative DW-MRI lesions (%) | 5 (14.3) | 28 (36.8) | 0.015 |
| Patients with at least one contralateral postoperative DW-MRI lesion (%) | 0 | 10 (13.1) | 0.03 |
| Patients with postoperative DW-MRI lesions and neurological symptoms (%) | 1/5 (20) | 12/28 (43) | 0.6 |
| Number of new lesions at DW-MRI | 10 | 85 | |
| Number of contralateral lesions at DW-MRI (%) | 0 | 19/85 (22) | 0.2 |
| Mean number of lesions per patient (range) | 2 (1–3) | 3.1 (1–8) | 0.1 |
Discussion
In these series of patients undergoing filter-protected carotid stenting through either a percutaneous femoral approach or a surgical cervical approach, we found a higher incidence of neurological events (19.8% vs. 2.3%) and a higher incidence of new DW-MRI ischaemic lesions (36.8% vs. 14.3%) after transfemoral percutaneous CAS. Based on our results, transcervical filter-protected CAS seems able to reduce both symptomatic and asymptomatic cerebral embolisation occurring during CAS, without carrying a risk of local complications. We indeed observed no ICA dissections or relevant cervical haematomas after this procedure. Even if mini-cervical access does not have a large clinical application for CAS, the few reports on this matter seem to confirm the very low incidence of access-related complications.16, 17 Particularly, no cranial nerve injuries were ever reported after transcervical CAS, thus showing the considerable difference of this cervical cut-down with the surgical incision needed for carotid endarterectomy.18 Finally, other authors have reported a low incidence of local complications after transcervical percutaneous CAS and haemostasis by a closure device.12 This approach would undoubtedly reduce the invasiveness and increase the appeal of this procedure but more data regarding its safety is needed to allow its more common use.
The results obtained in our series can be ascribed to instrumentation of the aortic arch which is lacking in the cervical procedure and is the only procedural difference between the two groups. Evidence regarding the impact of this phase on the total embolic load of transfemoral carotid stenting procedures has been provided by studies using either transcranial Doppler or DW-MRI. In unselected patients undergoing transfemoral carotid stenting Al-Mubarak et al. reported a mean number of 16 microembolic signals (MESs) during the first phase of the procedure, namely sheath placement, over a mean number of 68 MESs recorded during the whole procedure.19 This means that up to 23% of total procedural embolic load can be related to aortic arch instrumentation and CCA cannulation. Lower rates ranging from 10.2% to 11.4%, even though not negligible, can be extrapolated by data reported in other two transcranial Doppler (TCD) studies.20, 21 In an elegant DW-MRI study, Bendzus et al. documented the occurrence of new ischaemic lesions in 26% of the patients undergoing only diagnostic angiography of the supra-aortic trunks.7 A subgroup analysis of these results also showed an increased incidence in patients with history of vasculopathy (44%) and in patients with vessels difficult to probe (53%). Finally, DW-MRI studies reporting the incidence and the location of ischaemic lesions after transfemoral protected CAS have shown the occurrence of ischaemic lesions also in vascular territories not directly involved by the procedure.4, 6 While some authors have reported even 87% of contralateral lesions after transfemoral CAS,5 in our experience, the rate of contralateral lesions stopped at 22%. This is probably due to the treatment of many patients with complex anatomy by a cervical access. However, we found a significantly higher number of patients with contralateral lesions after transfemoral CAS (13%) than after transcervical CAS (0%).
All these data therefore confirm the considerable impact of aortic arch instrumentation on the procedural embolic load.
Current information coming from non-randomised clinical series argues for the beneficial use of cerebral protection devices in CAS.22 Several protection devices have been developed for transfemoral procedures but distal filters are undoubtedly the most widespread. In the few published articles on transcervical CAS, good results were reported in terms of neurological adverse events and postoperative DW-MRI ischaemic damage using ICA flow reversal as the cerebral protection system.23, 24 This system has the undeniable advantage of allowing cerebral protection before the first crossing of the carotid lesion; however, it requires CCA clamping and entails inter-hemispheric redistribution of cerebral blood flow. Intolerance to carotid clamping or flow reversal and relative complexity of arteriovenous fistula establishment may therefore explain the limited diffusion of this technique to date. On the other hand, while distal filters have the advantage of maintaining distal cerebral perfusion during the whole procedure, they do not provide protection during the first crossing of ICA stenosis. Nevertheless, the positive clinical results previously reported by Alexandrescu et al. after filter-protected transcervical CAS seem to be confirmed in our series.13 The technique used by Alexandrescu et al. also included transitory blood aspiration during first crossing of ICA stenosis to reduce embolic risk and yielded zero adverse neurological events in a group of 26 patients. Since patients with very tight stenoses, in whom crossing of the lesion may be difficult and risky, were excluded from this study and referred to surgery, we found this additional manoeuvre unnecessary and thus did not use it in any patient. However, we can speculate that blood aspiration or carotid clamping, in patients tolerating such manoeuvres, could have further decreased our low incidence of DW-MRI ischaemic lesions.
This study has some limitations that should be acknowledged. First of all, it is a non-randomised observational study including a relatively small number of patients. Second, we used DW-MR to compare the silent brain damage occurring after two different techniques but the real clinical impact of postoperative DW-MR lesions has yet to be clarified. This tool has nevertheless the advantage to allow an objective comparison between two techniques.
In conclusion, in this non-randomised single-centre study comparing surgical cervical access to percutaneous femoral access for CAS, we found cervical access to reduce the inherent cerebral embolisation of transfemoral carotid stenting. Even if an optimal cerebral protection system during transcervical CAS is still controversial, distal filters seem able to provide effective cerebral protection in selected patients. In our opinion, not all CAS procedures should be done by a cervical access but all interventionalists should acquire and be able to use this technique selectively, not only to overcome possible difficulties linked to femoral access but also to reduce adverse events in patients with complex anatomy.
Acknowledgements
None.
References
- . Current status in cervical carotid artery stent placement. J Cardiovasc Surg. 2009;50:29–37
- A comparison of carotid artery stenting with neuroprotection versus carotid endarterectomy under local anesthesia. Am J Surg. 2005;190:696–700
- Protected carotid-artery stenting versus endarterectomy in high-risk patients. N Engl J Med. 2004;351:1493–1501
- . New brain lesions after carotid stenting versus carotid endarterectomy: a systematic review of the literature. Stroke. 2008;39:1911–1919
- Silent cerebral ischemia detected with diffusion-weighted imaging in patients treated with protected and unprotected carotid artery stenting. Stroke. 2005;36:2389–2393
- 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
- . Silent embolism in diagnostic cerebral angiography and neurointerventional procedures: a prospective study. Lancet. 1999;354:1594–1597
- Carotid artery stenting is associated with increased complications in octogenarians: 30-day stroke and death rates in the CREST lead-in phase. J Vasc Surg. 2004;40:1106–1111
- . Target lesion ulceration and arch calcification are associated with increased incidence of carotid stenting-associated ischemic lesions in octogenarians. J Vasc Surg. 2008;47:88–95
- . The impact of increasing age on anatomic factors affecting carotid angioplasty and stenting. J Vasc Surg. 2007;45:875–880
- Aortic arch anomalies are associated with increased risk of neurological events in carotid stent procedures. Eur J Vasc Endovasc Surg. 2007;33:436–441
- . Choice of access for percutaneous carotid angioplasty and stenting. A comparative study on cervical and femoral access. J Cardiovasc Surg. 2009 May 19;[Epub ahead of print]
- Filter-protected carotid stenting via a minimal cervical access with transitory aspirated reversed flow during initial passage of the target lesion. J Endovasc Ther. 2006;13:196–204
- . How I do it: cervical access for carotid artery stenting. Am J Surg. 2006;192:779–781
- Characterisation of symptomatic and asymptomatic carotid plaques using high-resolution real time ultrasonography. Br J Surg. 1993;80:1274–1277
- Access site complications with carotid angioplasty and stenting. Surg Neurol. 2007;68:431–437
- . Transcervical carotid stenting with carotid artery flow reversal: 3-year follow-up of 103 stents. J Vasc Surg. 2007;46:864–869
- Laryngeal and cranial nerve involvement after carotid endarterectomy. Acta Otolaryngol. 2005;125:398–402
- . Effect of the distal-balloon protection system on microembolization during carotid stenting. Circulation. 2001;104:1999–2002
- Effect of two different neuroprotection systems on microembolization during carotid artery stenting. J Am Coll Cardiol. 2004;44:1966–1969
- Transcranial Doppler monitoring during stenting of the carotid bifurcation: evaluation of two different distal protection devices in preventing embolization. J Endovasc Ther. 2006;13:436–442
- Effects of age and symptom status on silent ischemic lesions after carotid stenting with and without the use of distal filter devices. Am J Neuroradiol. 2008;29:608–612
- . Transcervical carotid stenting with flow reversal protection: experience in high-risk patients. J Vasc Surg. 2007;46:49–54
- . Cerebral embolization during transcervical carotid stenting with flow reversal: a diffusion-weighted magnetic resonance study. Ann Vasc Surg. 2009;23:429–435
PII: S1078-5884(09)00578-4
doi:10.1016/j.ejvs.2009.11.011
© 2009 European Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
Volume 39, Issue 3 , Pages 252-257, March 2010
