Neurointerventional Report Downward Migration of Carotid Stent on 8 Months Follow-Up Imaging: Possible Stent “Watermelon- Seeding” Effect Aamir Badruddin, MD, Marc A. Lazzaro, MD, Muhammad A. Taqi, MD, Osama O. Zaidat, MD, MS From the Departments of Neurology (AB, MAL, OOZ, MAT), Neurosurgery and Radiology (OOZ), Medical College of Wisconsin, Froedtert Hospital, Milwaukee, WI. ABSTRACT BACKGROUND One method used to treat atherosclerotic carotid disease is carotid artery stenting (CAS). A rarely encountered limitation of this technique is stent migration. Here, we present a rare case of carotid stent downward migration found on follow-up imaging 8 months post operation. Keywords: Carotid stenting, stent migration, watermelon-seeding, restenosis, stent movement, stent displacement, stroke, TIA. Daniel M. Soref Clinical Neuroscience Fellow at Froedtert Hospital. METHOD/RESULT A 70-year-old man presented with aphasia and right-sided weakness secondary to highgrade (99%) proximal left internal carotid artery (ICA) stenosis that was treated with CAS. Stent apposition distally was achieved as well as proximally at the carotid bulb without crossing the bifurcation and without going distally beyond the ICA angulation to avoid kinking. Eight months follow-up computerized axial angiogram showed downward migration of the stent into the common carotid with restenosis distal to the stent. A 6—8 × 40 mm stent was deployed and the stenosed area restented with good results to overlap with the older stent and landed distal to the ICA angulation. Correspondence: Address correspondence to Osama O Zaidat, MD, MS, Vascular and Interventional Neurologist, Department of Neurology, Neurosurgery & Radiology, Director, Neurointerventional Program, Medical College of Wisconsin/ Froedtert West, 9200 W. Wisconsin Ave., Milwaukee, WI 53226. E-mail: szaidat@mcw.edu. CONCLUSION J Neuroimaging 2011;21:395-398. DOI: 10.1111/j.1552-6569.2011.00586.x Interventionalists should be aware of the rare possibility of migrating downward “watermelon-seeding” of carotid stents. This report may generate the hypothesis that the stent watermelon-seeding effect may be related to proximal placement of a short stent below the ICA angulation and at the carotid bulb in severely stenotic lesion. Introduction Angioplasty and stenting is a treatment option commonly presented for carotid artery stenosis.1 Stenting is routinely performed on patients deemed high surgical risk for carotid endarterectomy (CEA), and more recently has been shown to be equivalent to CEA in average surgical risk patients. 2−5 Neither CEA nor stenting approaches are without procedural risks. One potential complication is stroke from the embolic material set loose during angioplasty and stenting. Another associated risk with these procedures is hyperperfusion injury. A rare and an unexpected complication of these procedures is migration of a previously placed carotid stent. We submit a case of full migration of a carotid stent discovered upon examination of computerized tomographic (CT) axial neck angiographic images 8 months post operation. We also discuss potential and hypothetical anatomical and technical risks for carotid stent migration, procedural pitfalls, and possible methods of prevention. Case Report A 70-year-old Caucasian man with a history of coronary artery disease, hypertension, type II diabetes, hyperlipidemia, and glaucoma presented to the emergency department in December 2008, with complaints of transient right arm numbness and expressive aphasia lasting less than 1 minute. Transient ischemic attack (TIA) workup showed near occlusion of the left internal carotid artery just distal to the bifurcation. A string sign with distal reconstitution was also noted. He was admitted to the hospital for carotid revascularization. Initial noninvasive imaging revealed that the terminal portion of the lesion was not well visualized. Given his clinical comorbidities and the anatomical risk factor for endarterectomy, he was not a candidate for open surgical endarterectomy. He was brought to the angiosuite the following day for endovascular angioplasty and stenting of the symptomatic carotid lesion. Digital subtraction angiography (DSA) confirmed the presence of string sign of the left internal carotid artery. Stenosis was measured at 99% by NASCET criteria. The length of the stenotic segment was approximately 12 mm and the width of the nondiseased segment just distal to the lesion was 4.6 mm (Fig 1A). All measurements were preformed using the software provided on the biplane DSA machine. Prestenting balloon angioplasty was performed using 3 mm × 20 mm Sterling balloon (Boston Scientific, Natick, MA) inflated at 6 atmosphere pressure. A 5 mm Emboshield (Abbott Laboratories, Abbott ◦ Copyright C 2011 by the American Society of Neuroimaging 395 Fig 1. (A) Initial carotid angiogram demonstrating high-grade stenosis (solid arrow) and carotid angulation (dashed arrow). (B) Carotid angiogram status post initial stenting, note stent edges and wall apposition (arrows indicate the distal and proximal initial stent landing zones). (C) Follow-up angiogram demonstrating downward stent migration (solid arrow) and distal restenosis (dashed arrow). (D) Native angiogram post placement of second stent. Note the placement of considerably longer tapered stent past the angulation of the internal carotid artery (dashed arrow) and proximal landing zone in the common carotid artery overlapping the initial stent (solid arrow). (E) Final digital subtracted carotid angiogram showing no residual stenosis with minimal kinking at the distal margin of the new second stent. Park, IL) distal protection device was then deployed. Subsequently, a 7 mm ×20 mm Xact stent (Abbott Laboratories, Abbott Park, IL) was deployed across the lesion with good proximal and distal margins and arterial wall apposition, at the edge of the distal area of the stenosis (Fig 1B). The stent was deployed proximal to the angulation of the internal carotid artery, 396 Journal of Neuroimaging Vol 21 No 4 October 2011 due to potential complication of vessel kinking at the distal edge of the stent from insertion of a stiff endoluminal device. A final poststent angioplasty was performed using a 6 mm × 20 mm Sterling balloon. No residual stenosis was noted. On final imaging, stent placement was apposed to the vessel wall and spanned the stenotic carotid segment (Fig 1B). The patient was discharged home the following day on clopidogrel and aspirin. He remained symptom free, however, CTA performed 8 months following the initial procedure demonstrated proximal downward migration of the stent, resulting in a major segment of the stent length laying in the common carotid artery. Furthermore, there was an 80-90% restenosis of the left ICA distal to the migrated stent at the prior area of stenosis (Fig 1C). Two weeks after the follow-up CTA, repeat angioplasty with restenting was performed. An initial left carotid DSA confirmed the restenosis of the left internal carotid artery distal to carotid bulb and estimated diameter narrowing of 80%. The stent had migrated downward as noted on the CTA when compared to the initial angiogram and stent placement (Fig 1B) A tapered 6-8 mm × 40 mm Xact stent was then advanced across the stenotic lesion and deployed successfully with distal embolic protection device using FiberNettm 3.5-5 mm (Invatec, Roncadelle, Italy). The stent, at the distal ends of the stenosis and common carotid artery, was situated such that it spanned above the angulation, with good landing zone, and proximally in the distal common carotid artery. The placement of the new stent overlapped the distal portion of the previously placed stent. Poststent balloon angioplasty was performed using a 6 mm × 20 mm Sterling balloon. The distal embolic protection device was removed and no residual stenosis was noted on the followup carotid angiogram. A significant improvement in blood flow was noted in the left anterior intracranial circulation. The patient was discharged home the next day on clopidogrel and aspirin. Discussion Descriptions of balloon watermelon-seeding effect have demonstrated that the balloon is unstable across a stenosis and will subsequently “slip” out of the stenotic segment toward the area of least constraint. Similarly, a stent may also “migrate” away from a severely stenotic lesion toward an area of artery that allows its full expansion.6,7 This effect is not as commonly seen with stents. Migration of carotid artery stent is a rare occurrence. Our review found only one reported case of ICA stent migration.6 Technical Factors One of the theoretical risk factors for stent migration is under sizing the stent. In our patient, the vessel size was 5.5 mm at its broadest proximal segment and the stent used was 7 mm in diameter. Therefore, the original stent was not significantly undersized. However, a hypothetical explanation with the selected stent size to the stent migration is that the original lumen caliber size that was initially reduced due to low flow, reexpanded with subsequent dynamic vessel wall expansion after establishing the flow, therefore, rendering the stent small for the newly reexpanded vessel. Another possible risk factor for stent migration is deployment of a short stent over a calcified lesion, in the current case; lesion length was 12 mm. An attempt was made to stay above the bifurcation and below the carotid angulation in a 12-mm stenotic lesion. This set of circumstances led to the selection of a 7 mm × 20 mm stent length, which ultimately may not have provided enough vessel wall support in the proximal segment, leading to the distal segment of the stenosis to be barely covered, which may have potentially caused the stent to migrate. Additionally, the stent/vessel diameter ratio was greater in the distal segment, and this may have created an outward radial force. This force may have been downwardly redirected, shifting the stent, as there was less resistance at the proximal end of the carotid bulb. Due to the circumstances of proximally anchoring a short stent within a relatively confined landing zone, the stent may have migrated downward. Prevention of migration may have occurred if a longer stent was placed with a more proximal landing zone in the common carotid artery. The use of a closed stent with greater metal surface area may have possibly allowed for more endothelial to metal contact, and thus more anchoring thereby preventing migration, however, there is no literature to our knowledge on stent migration prevention, which supports open versus closed cell design. Lesion Factors The lesion treated had near occlusion and string sign. It is possible that such lesions impart greater risk for stent migration. The case of downward stent migration may have resemblance to the watermelon-seeding effect that is commonly seen with balloon angioplasty of severe stenosis.7 Furthermore, as demonstrated in our patient, any stent migration is expected to move in the direction of blood flow and not against it. Endothelialization of the stent begins shortly after placement and therefore it is reasonable to believe that in few weeks the stent struts would be enmeshed with the vessel wall, making any migration or displacement difficult. Therefore, the migration of the stent is likely to happen early after its deployment. In our patient, the first follow-up imaging was done 8 months post treatment, as the patient was lost to follow-up due to noncompliance. Therefore, it is difficult to pinpoint the exact timeframe when stent migration occurred. However, given the explanation above, migration may have occurred in the first few days after the procedure. Few reports of migration of the self-expanding intracranial stents such as Neuroform and Enterprise stents have been published.8−12 This may also be related to oversizing at the distal end while relative or just exact/smaller sizing proximally, seen similarly in our case. However, the migration of carotid stent is a rare occurrence, and to our knowledge, has only been reported once.5 In the case presented here, we believe that the longer stent length with adequate landing zone, as well as poststenting balloon angioplasty, would make stent migration less probable. The poststenting balloon angioplasty would hypothetically provide mechanical anchoring of the stent struts to the calcified carotid plaque, and stabilize the stent. In conclusion, this case is instructional on the potential mechanism and risk factors of downward migration, or delayed stent watermelon-seeding effect. It is helpful to note that placing a nontapered stent across a large differential gradient in vessel diameter proximally and distally may lead to stent watermelonseeding effect. Also, early imaging follow-up should be obtained, perhaps even in 2-4 weeks to assess the patency and Badruddin et al: Downward Migration of Carotid Stent 397 station of the deployed stent. The time period for greatest risk of stent migration postprocedure is unknown, and would be a relevant question to answer with future studies. The authors wish to thank Sara Berentsen for her help with English editing. References 1. Yadav JS, Wholey MH, Kuntz RE, et al. Protected carotid-artery stenting versus endarterectomy in high-risk patients. N Engl J Med 2004;351(15):1493-1501. 2. SPACE Collaborative Group,Ringleb PA, Allenberg J, Bruckmann H, et al. 30 day results from the SPACE trial of stent-protected angioplasty versus carotid endarterectomy in symptomatic patients: a randomised non-inferiority trial. Lancet 2006;368(9543):1239-1247. 3. CAVATAS Investigators. Endovascular versus surgical treatment in patients with carotid stenosis in the Carotid and Vertebral Artery Transluminal Angioplasty Study (CAVATAS): a randomised trial. Lancet 2001;357(9270):1729-1737. 4. Mas JL, Chatellier G, Beyssen B, et al. Endarterectomy versus stenting in patients with symptomatic severe carotid stenosis. N Engl J Med 2006;355(16):1660-1671. 398 Journal of Neuroimaging Vol 21 No 4 October 2011 5. Brott TG, Hobson RW 2nd, Howard G, et al., CREST Investigators. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med. 2010;363(1):11-23. Epub 2010 May 26. 6. Castriota F, Martins EC, Dall’Olmo CA, et al. Stent migration as a late complication following carotid angioplasty and stenting. EuroIntervention 2008;4(3):397-399; discussion 400-404. 7. Alfonso F, Perez-Vizcayno MJ, Gomez-Recio M, et al. Implications of the “watermelon seeding” phenomenon during coronary interventions for in-stent restenosis. Catheter Cardiovasc Interv 2005;66(4):521-527. 8. Kelly ME, Turner RD 4th, Moskowitz SI, et al. Delayed migration of a self-expanding intracranial microstent. AJNR Am J Neuroradiol 2008;29(10):1959-1960. 9. Lavine SD, Meyers PM, Connolly ES, et al. Spontaneous delayed proximal migration of enterprise stent after staged treatment of wide-necked basilar aneurysm: technical case report. Neurosurgery 2009;64(5):E1012. 10. Lee TH, Kim DH, Lee BH, et al. Preliminary results of endovascular stent-assisted angioplasty for symptomatic middle cerebral artery stenosis. AJNR Am J Neuroradiol 2005;26(1):166-174. 11. Thomas JA, Watson VE, McGrail KM. Surgical management of a paraclinoid aneurysm containing a displaced neuroform microstent: technical case report. Neurosurgery 2008;63(4):E817E818. 12. Yahia AM, Gordon V, Whapham J, et al. Complications of Neuroform stent in endovascular treatment of intracranial aneurysms. Neurocrit Care 2008;8(1):19-30.