In the aging population, stroke is the most common and disabling neurological disorder, with more than 500,000 annual strokes in the United States.1 Carotid artery stenosis is a significant risk factor for stroke. Surgical treatment for carotid artery stenosis has been the traditional standard of care.2 In the North American Symptomatic Carotid Endarterectomy Trial (NASCET), it was documented that carotid endarterectomy (CEA) is beneficial in reducing stroke risk in those patients with significant stenosis.3,4 CEA has a significant perioperative morbidity and mortality rate, the risk of which depends on the skill and experience of the surgeon and staff.3,4 Carotid artery stenting (CAS) is a nonsurgical way of unblocking the atherosclerotic carotid arteries. Embolic protection prevents particles that become dislodged during CAS from moving to the brain where they can cause stroke or death. Two recently completed studies, ARCHeR 2 and SAPPHIRE, compared CEA with CAS utilizing embolic protection in high surgical risk patients.5,6 In these trials, CAS utilizing embolic protection was found to be just as effective as, and much less invasive than, surgery in preventing stroke. The SAPPHIRE Trial was the first randomized study to compare CAS with embolic protection to CEA. This trial showed favorable results for the new procedure, determining the noninferiority of carotid artery angioplasty and stenting to carotid endarterectomy in the treatment of carotid artery stenosis.6 The PROTECT Trial, sponsored by ev3 Inc. (Plymouth, Minnesota) for CE Marking, evaluated the performance requirements of the SPIDER™ Embolic Protection Device in patients who were candidates for percutaneous transluminal angioplasty (PTA) and/or stenting. Qualified patients were treated with distal protection using the SPIDER Embolic Protection Device. Methods Patient selection. The PROTECT Trial enrolled 74 patients 18 years of age or older between November 2001 and September 2002 across five European study centers in three countries: Belgium, Germany and Italy. The appropriate local ethics committee approved the research protocol and informed consent. Written informed consent was obtained from all patients prior to enrollment. Table 1 summarizes the study parameters, including patient selection criteria, study design and patient enrollment. Qualified patients presented with carotid artery stenosis suitable for femoral access percutaneous transluminal angioplasty (PTA) and/or stenting with distal protection. The patients included were either asymptomatic with a ? 70% lesion occlusion, or were symptomatic with a ? 50% lesion occlusion. The normal parallel portion of the internal carotid artery distal to the stenosis where the Capture Wire filter was placed was between ? 2.0 mm and ? 7.0 mm in diameter. Patients with a total occlusion of the target vessel during the pre-procedure angiography were not included. Patients who experienced an acute stroke within 14 days prior to the procedure, as well as patients with a major residual neurologic deficit (stroke scales: Barthel ? 60, NIH ? 15, or Rankin > 3) in a previously defined territory at pre-procedure neurological exam were excluded. Any patients with intolerance to heparin, aspirin or clopidogrel were also excluded. A complete list of exclusion criteria is presented in Table 2. System description. The ev3 Inc. SPIDER Embolic Protection Device was evaluated in the PROTECT Trial. The SPIDER System is a percutaneously delivered distal embolic protection filter designed to capture and remove debris from internal atherosclerotic carotid artery vessels during interventional procedures. The SPIDER System is the only embolic protection device that allows use of the physician’s preferred guidewire to initially cross the lesion. The SPIDER System consists of a Capture Wire (filter), delivery catheter, stylet, and recovery catheter. All components are biocompatible, packaged sterile for single-use only, and are compatible with most standard interventional devices. The Capture Wire consists of a 0.014 inch stainless steel wire and a nitinol mesh filter mounted on the 175/320 cm polytetrafluoroethylene-coated stainless steel wire. The nitinol wire used in the filter measures 0.0013 inches. The inlet and outlet dimensions of the marker band on the filter are 0.01 and 0.026 inches, respectively. A gold loop at the opening of the filter mouth provides a visual marker indicating the opening of the filter mouth. The gold loop is made of nitinol with a gold tungsten coil. The SPIDER System is available in 5 filter sizes: 3–7 mm. The delivery catheter is used to exchange the physician’s preferred guidewire with the Capture Wire, and deploy the filter at the desired location. The stylet creates a transition from the delivery catheter to the primary guidewire. The recovery catheter is used to recover the filter and captured atherosclerotic debris from the carotid artery vessel. The recovery system dimension is 4.2 French for the 3, 4 and 5 mm filters and 4.9 French for the 6 and 7 mm filters. Three figures depict the Spider embolic protection device. Figure 1 is a schematic identifying the delivery catheter, stylet and preferred guidewire. Figure 2 is a picture of the SPIDER System components. Figure 3 depicts the Spider device prior to and following deployment of the filter in the lumen of the carotid artery vessel. Study endpoints. Performance was assessed by determining the incidence of major adverse neurological events (MANE) through 30 days postprocedure, along with the successful placement and recovery rates of the filter. MANE was defined as stroke or death. Stroke was further defined as a new focal neurological deficit of presumed vascular origin. When classifying strokes as major or minor, the Clinical Events Committee (CEC) took into consideration the NIH Stroke Scale7 and the Rankin Stroke Scale8 to differentiate and classify each stroke. Table 3 presents the guidelines used by the CEC to classify major and minor strokes. Technical success was defined as placement of the filter at the distal edge of the lesion, placement of the recovery catheter and successful retrieval of the filter with the recovery catheter. An inability to place or retrieve the filter was concluded to be a SPIDER Distal Protection Device failure. Analysis of the contents of the filter component of the device was performed to determine the presence and nature of embolic material captured within the filter. Procedure for SPIDER device placement. Selective angiography of the carotid artery was performed to identify the anatomical characteristics of the intracranial and carotid vasculature and to best isolate and define the lesion in two orthogonal views. Based on the North American Symptomatic Carotid Endarterectomy Trial (NASCET) method, calculations were performed to determine the degree and length of the stenosis, as well as the distal normal internal carotid artery diameter. Using standard procedures for carotid stenting, a guide catheter was introduced in the femoral artery and placed in the high common carotid artery. Patients received a minimum of 300 mg of aspirin orally or the equivalent intravenous dose prior to the procedure. In addition, patients received clopidogrel 75 mg for 4 days prior to the procedure or a bolus of 300 mg prior to the procedure. Heparin was administered as needed to maintain an ACT of ? 250 seconds throughout the interventional portion of the procedure. If abciximab was administered prior to the start of the procedure, an ACT of > 200 seconds was maintained. The Instructions for Use for the SPIDER device were followed for preparation, placement and recovery. Following evaluation of the region in the target vessel where the filter was to be deployed, the correct system size was chosen from the five available sizes. The Capture Wire filter was placed distally to the target lesion with the help of the delivery catheter/stylet positioned at least 4 to 5 cm distal to the site of the treatment. The position was controlled under fluoroscopy with the radiopaque marker band on the distal tip of the delivery catheter and the proximal radiopaque marker of the filter. Deployment of the filter was achieved when the radiopaque proximal end of the filter was distal to the treatment lesion and the radiopaque loop at the mouth of the filter was visible. Redeployment or removal of the filter could be performed at this step to achieve proper filter position at least 1.5 cm distal to the lesion being treated. Once the filter was correctly deployed in the desired location, the delivery catheter was removed and the interventional procedure performed. At the end of the procedure, the filter was removed with the use of the recovery catheter. Histologic processing of atherosclerotic debris collected. The filters containing atherosclerotic debris were shipped upright in 15 ml conical tubes containing 10 ml of 10% neutral-buffered formalin. The specimens were removed from the collection tube and photographed under a dissection microscope. The material within the baskets was teased away with fine forceps and placed in histologic cassettes. The formalin in the shipping container was also filtered through lens paper and processed for histology in the event debris dislodged from the basket during shipping. The samples were dehydrated in a graded series of alcohols and embedded in paraffin. Approximately 8 sections were cut at 4 µm intervals and mounted on charged glass slides. The first 2 slides were stained with hematoxylin-eosin (H & E) and Movat pentachrome stains for histologic identification of the type of debris. Immunohistochemical stains were performed when necessary for the positive identification of macrophages (CD68), T-lymphocytes (CD45RO), platelets (CD61), fibrin (T2G1) and smooth muscle cells (a-actin). Evaluation of plaque debris using light microscopy. Semi-quantitative assessments of acellular and cellular plaque components were performed on H & E stained sections; Movats were used for verifying collagen, elastin and proteoglycans. Samples were considered to have maximal tissue if some atherosclerotic material was present in every high-power light microscopy field, moderate if every other field showed material, minimal when only sparse material was present, and extremely minimal if essentially only one or two small particles were present. The following material was identified as plaque: macrophage foam cells, necrotic core, cholesterol clefts, collagen, platelets, fibrin and smooth muscle cells. Measurement of particle size. Measurement of particle size was determined using computer software with automated planimetry (BioQuant, R&M Biometrics, Inc., Nashville, Tennessee). The system software was calibrated using a stage micrometer. Only the largest particles from each case were measured with no more than 20 particles from each specimen analyzed. The operator manually assigned the actual areas and lengths while the measured outputs were computer-generated. The total area of the captured particles and the major and minor axis of each particle were measured. Results A total of 74 patients were enrolled. There were 52 males (70%) and 22 females, and the mean age was 72 years (± 2 years). The medical history that may have classified patients as high risk in the PROTECT Trial are summarized in Table 4. Of the 7 listed risk factors, 45% (33/74) of enrolled patients had at least 1, 14% (10/74) of patients had 2, and 3% (2/74) of the patients had 3 or 4 risk factors. The PROTECT Trial enrolled symptomatic patients with lesion occlusion ? 50%. Asymptomatic patients were eligible as long as the lesion occlusion was ? 70%. Thirty-day MANE rate. One patient withdrew from the study prior to completion of the one-month follow-up, and therefore the 30-day MANE rate is based on 73 patients who completed the 30-day follow-up visit. There were a total of 5 MANE events in 5 patients reported through 30 days postprocedure, 1 major stroke and 4 minor strokes. There were no deaths. The total 30-day MANE event rate was 6.8% (5/73; 95CI:2.3%, 15.3%). One major stroke and 2 minor strokes occurred during the procedure hospital stay. The remaining 2 minor strokes were reported following hospital discharge. There were a total of 6 serious adverse events (SAE) reported by 6 patients. The overall SAE rate was 8.2% (6/73) and is summarized in Table 5. An SAE was defined as an event that resulted in the hospitalization or prolongation of hospitalization due to potential disability or due to danger of life that resulted in an intervention or that was terminal. Two events were procedure-related. One patient experienced ischemic damage to the optic nerve. The second patient was treated for right carotid artery angioplasty and stenting without incident. One day following the procedure, the patient was hospitalized for a prolonged period due to occlusion of an arterial branch in the retina of the left eye. The patient received rheological treatment over the course of 7 days in combination with heparin which produced an improvement of the visual disturbance. Four events were not device or procedure-related. Two patients experienced prolonged angina pectoris, 1 patient was hospitalized 6 days postoperative with an aneurysmal hematoma of the medial side of the right thigh (PTA entry site) with extension to the right hypogastric region. The aneurysmal hematoma was treated with a thrombin injection. One patient with a gastric polyp was hospitalized with acute enteritis and dyspepsia. The patient was treated and the event resolved without incident. Filter placement and retrieval. All 74 patients enrolled in the study had successful placement of the Spider device. Eight patients required a second attempt with a new Spider device to successfully deliver the device and 1 patient required a third attempt for successful placement. One lesion per patient was treated following placement of the device. The additional attempts were required in 9 patients due to device damage during the delivery procedure such as bending of the guidewire or damage to the filter itself as it was pushed through the delivery catheter. It is possible that anatomical considerations and operator expertise may have contributed to the need for additional attempts to successfully place the device. The fact that all patients had successful device placement, 65 patients after the first attempt, 8 additional patients after the second attempt, and 1 additional patient after the third attempt, suggests a possible learning curve necessary to successfully place the device. Counting each filter placement attempt, there was an overall success rate of 88.1% (74/84). A summary of filter placement success is presented in Table 6. The SPIDER device filter retrieval rate was 98.6% (73/74). One patient required a smaller retrieval catheter to successfully remove the filter without incident. Pathology data from collected debris. Debris was collected in 79% (58/73) of the filters evaluated. One filter was not returned for analysis. Atherosclerotic debris was collected in 67% of the filters (49/73), while an additional 9 contained platelets and/or fibrin in the absence of atherosclerotic debris. Of the 49 devices with captured atherosclerotic debris, 29% (14/49) demonstrated macrophage foam cells, 61% (30/49) smooth muscle cells, 76% (37/49) necrotic core, 37% (18/49) cholesterol clefts, 65% (32/49) collagen/elastin, and 78% (38/49) platelets/fibrin. The mean values for minimal and maximal particle area were 0.01 mm2 and 0.31 mm2, respectively. The average particle size was 0.06 mm2. Mean maximal particle length was 0.33 mm; mean minimal length was 0.18 mm. A frequency distribution of particles grouped by area showed that the majority of devices contained particles ? 0.1 mm2. For major particle length, the majority of devices had particles between 0.1 and 0.4 mm. The minor lengths were between 0.1 and 0.2 mm. Discussion This study demonstrated performance of the SPIDER Embolic Protection Device when used in patients presenting with carotid artery stenosis suitable for percutaneous transluminal angioplasty (PTA) and/or stenting. The incidence of major adverse neurological events defined as stroke or death was low, at 6.8% (5/73). This 30-day event rate is comparable to the recently completed ARCHeR 2 nonrandomized study that compared CEA with carotid artery stenting (CAS) and distal protection in 278 high surgical risk patients. The 30-day MANE rate in this study was also low, at 6.8%. The ARCHeR 2 trial featured the Acculink™ carotid stent and the Accunet™ embolic protection device (Guidant Corporation, Santa Clara, California) designed specifically for use in the carotid artery. The difference in the stroke or death MANE rates between PROTECT and ARCHeR 2 was not significant. This was determined by using a chi-square test for the PROTECT-ARCHeR 2 comparison. The p-value for the comparison was 0.9964 (Table 7). No deaths or myocardial infarctions occurred in the PROTECT Trial. The incidence of other procedure-related SAEs was also low, at 2.7% (2/73). The overall SAE rate, including nondevice or procedure SAEs, was low, at 8.2% (6/73). The overall success rate for placement of the filter at the distal edge of the lesion was 88.1%. Some patients required a second attempt for successful placement, with 1 required a third attempt. It is possible that anatomical considerations or operator expertise may have contributed to the 9 unsuccessful first attempts. Evidence of the need for embolic protection during carotid stenting procedures also was confirmed with the observation that the SPIDER device successfully collected atherosclerotic plaque, platelets and/or fibrin in 79% of the cases. Conclusion The PROTECT Trial demonstrated that the performance of the SPIDER Embolic Protection Device in the carotid artery complied with the relevant product directives for CE Marking. The PROTECT Trial successfully demonstrated the performance of the SPIDER device in patients who were candidates for percutaneous transluminal angioplasty by determining the incidence of major adverse neurological events through 30 days postprocedure. The trial utilized the SPIDER device in conjunction with PTA and stenting. The incidence of MANE was low and comparable to what was reported in another recently-completed distal protection study. The device was successfully placed distal to the carotid artery lesion and removed without incident. In addition to the SPIDER Embolic Protection Device, the rapid exchange version is now commercially available in Europe and select Latin American and Asia Pacific countries as the SPIDERx™ embolic protection device.
1. American Heart Association. Heart and Stroke Facts Statistical Supplement. New York: American Heart Association. 1994, p 12. 2. Debakey MH. Carotid endarterectomy revisited. J Endovasc Surg 1996;3:4. 3. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high grade carotid stenosis. N Engl J Med 1991;32:445‚Äì453. 4. Paciaroni M. Medical complications associated with carotid endarterectomy. NASCET Trial. Stroke 1999;30:1759‚Äì1763. 5. Gray WA. ARCHER trials: Final one-year results. American College of Cardiology Annual Scientific Session, March 7, 2004, New Orleans, Louisiana. 6. Yadav JS, Wholey MH, Kuntz RE. Protected carotid artery stenting versus endarterectomy in high-risk patients. N Engl J Med 2004;351:1493‚Äì1501. 7. Lynden P, Brott T, Tilley B, et al. Improved reliability of the NIH Stroke Scale using video training. NINDS TPA Stroke Study Group. Stroke 1994;25:2220‚Äì2226. 8. Van Swieten J, Koudstaal PJ, Visser MC, et al. Interobserver agreement for the assessment of handicap in stroke patients. Stroke 1988;19:604‚Äì607.