Introduction
Carotid artery angioplasty and stenting (CAS) has gained acceptance in the past decade as an alternative strategy for management of carotid artery stenosis. The recent SAPPHIRE trial¹ has shown that, in high surgical risk patients, CAS is at least as good if not superior to carotid endarterectomy. The initial concern over high thromboembolic complications has been addressed by the introduction of distal embolic protection devices (EPD). Early experience with EPD indicates that they reduce microemboli-related strokes during CAS.² However, deployment of the device increases the risk of complications and the complexity of the CAS procedure. We present a case of CAS complicated by filter detachment and localized dissection, treated with mechanical filter retrieval and restenting.

Clinical Case
An 83-year-old gentleman was referred to our center for management of symptomatic carotid artery stenosis manifested by transient ischemic attacks (TIA). He reported intermittent episodes of right lower extremity weakness lasting 10–15 minutes. He had a history of hypertension, deep vein thrombosis (for which he was on chronic warfarin therapy) and advanced chronic obstructive pulmonary disease (COPD), requiring chronic oxygen supplement and limiting his functional ability to walking 25 feet before developing dyspnea. His exam was notable for left-sided carotid bruits, but revealed no focal neurologic deficits.

Due to the patient’s age and advanced COPD, he was considered to be at high risk for carotid endarterectomy (CEA), and therefore it was decided to proceed with percutaneous carotid intervention. A 5 Fr Vitek catheter (Cook Inc., Bloomington, Indiana) was used to engage selectively in the carotid arteries. Carotid angiogram confirmed high-grade stenosis of the left ICA (Figures 1 and 2). The lesion was in a tortuous segment of the ICA with a 90-degree angulation. The ICA/CCA angle appeared to be moderate (about 45 degrees). The Vitek catheter was exchanged over a TAD II wire (Mallinckrodt, St. Louis, Missouri) for 7 Fr Flexor shuttle sheath (Cook Inc.). An Accunet Embolic Protection filter wire (Guidant Corp. Santa Clara, California) was placed distally in the ICA without difficulty, and the left internal carotid lesion was pre-dilated with a Quantum 4.0 x 15 mm balloon (Boston Scientific, Maple Grove, Minnesota). The balloon was exchanged for an Acculink Rx 8-6 x 30 mm self-expanding stent (Guidant Corp.). Since the lesion was mainly in the internal carotid artery, the stent was specifically placed just in the internal carotid rather than extending it into the common carotid artery (CCA). The stent was further dilated with a 5.0 x 15 mm balloon. When removal of the filter was attempted, the retrieval sheath became snagged on the stent struts and could not be passed past the proximal end of the stent (Figure 3). Therefore, further dilation with a 6 x 20 mm UltraSoft balloon (Boston Scientific) to create a different angulation and allow advancement of the sheath was attempted, but was unsuccessful. Despite multiple catheter manipulations and rotations, the retrieval sheath could not be advanced. At this point, it was decided to attempt to pull the filter back into the Shuttle catheter without the retrieval sheath. While pulling back, the filter became snagged on the stent and broke off, leaving a stent and filter apparatus in the common carotid artery (Figure 4). Consideration was given to stenting the whole apparatus against the CCA wall, but it was felt that the apparatus might be bulky enough to prevent adequate stent opposition or cause significant flow limitation. Therefore, we decided to attempt mechanical capture and removal of the apparatus. Attempts at removal using a 2–4 Ensnare Minisnare (Medical Device Technologies, Gainesville, Florida) were unsuccessful. Next, a three-pronged grasper device (Applied Medical, Rancho Santa Margarita, California) was used to grab and pull out the stent-filter apparatus (Figure 5). Following this, a fairly pronounced dissection was noted at the site of the previous lesion (Figure 6), attributed to the stent-filter apparatus trauma to the vessel wall on pulling back. The dissection was treated with placement of an 8-6 x 40 mm stent in the ICA, extending into the CCA. Despite the difficulty encountered during removal of the initial EPD device, we opted to deploy a new filter system prior to the second stent placement because of the concern over microembolization from the dissection site. We felt confident that extending the stent in the CCA would lessen the angulation and allow straight passage of the removal sheath through the stent. After stent deployment, no further balloon dilation was necessary, and the filter was easily removed with a filter removal sheath (Figure 7). Throughout the procedure, the patient’s neurologic exam continued to be normal. He had no symptoms of TIA or stroke over the next 24 hours, and was discharged the following day. At three-month follow-up he was symptom-free, with no further recurrence of his baseline TIAs.

Discussion
Since the first reported case of carotid artery angioplasty by Kerber et al. in 1980,³ the interest in percutaneous carotid intervention has risen steadily because of the perceived benefits of less invasiveness and faster convalescence as compared with carotid endarterectomy (CEA).^4,5

Concerns over the potential for distal embolization of plaque fragments to the brain has generated great concern regarding the safety of CAS. Cerebral protection devices were designed to be placed distal to the stenosis and collect the embolic debris before it enters the intracranial circulation. The use of EPD during CAS can significantly reduce the occurrence of thromboembolic complication rates, namely, strokes. Kastrup et al.² conducted a systemic review of 40 studies without EPD, and 14 studies with EPD. The combined stroke and death rate within 30 days was 1.8% in patients treated with cerebral protection devices compared with 5.5% in patients treated without cerebral protection devices (p < 0.001). This effect was mainly due to a decrease in the occurrence of strokes (4.8% without cerebral protection versus 0.8% with cerebral protection; p < 0.001), as the death rate was nearly identical. The authors estimated that there was a three-fold increased risk of any stroke or death and a greater than six-fold increase of minor stroke within 30 days of CAS without protection as compared to with protection.

Recent technical refinements of EPDs have minimized complexity of device deployment, decreased complications and ultimately led to the widespread adoption of cerebral protection during CAS.
In general, EPDs are deployed with a high success rate of over 97%.^6–8 Internal carotid tortuousity can impede proper placement of a cerebral protection device or render retrieval of the system difficult. Severe angulation is more common in patients over 80 years old and associated with increased complication rate.⁹ Carotid tortuosity indices were advised by Lin el al.¹⁰ to reflect increasing complexity in wire and catheter passage. The grading is as follows: vessel with < 30 degree angulation from the centerline flow of blood (Grade I), 30–60 degree angulation (Grade II), and > 60 degree angulation (Grade III).

Complications related to EPD are not uncommon. In one study,⁷ thromboembolization during initial lesion crossing by an EPD occurred in 0.79% of cases, distal internal carotid artery spasm in 3.6% and flow impairment not related to spasm in 7.2% (generally resolving after filter removal).

Difficulty with retrieval of the filter through the stent is relatively uncommon and is often underreported across the studies. Reimers et al.⁷ reported difficulty with EPD removal in 8.4% of their cases, but the actual rate, in our experience, has been much lower, in the 1% range. In the presented case, the likely etiology for the encountered difficulty is a combination of the high-profile, nontapering end, and lack of flexibility of the filter retrieval sheath (Figure 8). In addition, limiting the stent to the internal carotid artery and the significant angulation at its origin made it impossible to pass the sheath without its getting snagged on the stent struts. In most instances, additional stent dilation changes the angulation and allows sheath advancement.

Placement of a second wire, or “body wire,” is another maneuver that can be attempted to straighten the vessel. Alternatively, a different catheter can be used to capture the filter. For example, a multi-purpose coronary catheter, which has a different tip profile and flexibility, can be utilized. Following the second filter deployment, extending the stent into the common carotid artery enabled us to overcome the anatomical difficulty and maneuver the retrieval sheath up through the stent.

In conclusion, certain anatomic characteristics like tortuosity and angulation are more common in octogenarians and may impact the technical difficulty and procedural risk of cerebral protection system deployment. However, with accumulating experience and the anticipated technical improvements in EPD, it is likely that current complication rates will decline in the near future.

Commentary

“Management of Detached Accunet Embolic Protection Filter During Percutaneous Carotid Artery Intervention” by Chane et al. is a very important paper, as it describes something that everyone involved in carotid intervention needs to know and understand: these “wonderful new technologies…” are not perfect, and trouble is always around the corner! While we all recognize that distal embolic protection is probably an important and necessary component in every CAS procedure, use of such devices does add — unquestionably — a layer of complexity and potential pitfall. Through seasoned skills and a bit of luck, the authors managed to avoid catastrophe, but it could have easily gone the other way. VDM readers will do well to think through these issues, and read this article more than once! “Unbounded enthusiasm” for CAS needs to be tempered by the realization that potential trouble looms large, and the fact that there is a brain just beyond your wire!

Frank J. Criado, MD
frank.criado@medstar.net