Author + information
- Received December 27, 2001
- Revision received April 25, 2002
- Accepted May 24, 2002
- Published online September 4, 2002.
- ↵*Reprint requests and correspondence:
Dr. Joachim Schofer, Center for Cardiology and Vascular Intervention, Othmarscher Kirchenweg 168, 22763 Hamburg, Germany.
Objectives We sought to prospectively assess the feasibility and in-hospital efficacy of the PercuSurge GuardWire temporary balloon-occlusive system for neuroprotection during carotid angioplasty and stenting (CAS).
Background Carotid angioplasty and stenting harbors a risk of distal embolization. Cerebral protection devices are currently under clinical investigation.
Methods Ninety-six consecutive patients with carotid bifurcation disease underwent a total of 102 CAS procedures with the intention to use the GuardWire for neuroprotection.
Results GuardWire deployment was achieved in 99 procedures performed in 93 patients (97%). Device failure (n = 3) and severe neurologic responses to balloon occlusion of the targeted carotid artery (n = 2) accounted for five additional procedures that were essentially concluded without neuroprotection, for a total of 94 procedures completed as intended in 88 patients (92% procedural feasibility rate). Carotid angioplasty and stenting was performed successfully in 94 patients (100 procedures). There were no in-hospital deaths; but three patients (3.1%) sustained strokes, and two patients experienced transient ischemic attacks, for a total periprocedural complication rate of 5.2%. One major stroke occurred with the GuardWire in place, whereas two minor strokes were observed in patients in whom the device could not be deployed. Thus, successful neuroprotected CAS without major neurologic events was achieved in 87 patients (91%).
Conclusions The GuardWire temporary balloon-occlusive system is feasible as an adjunct to CAS in the majority of patients. It is associated with a 3.1% rate of major periprocedural neurologic complications. Adverse neurologic reactions to balloon occlusion may prohibit effective use of the system in about 2% of patients.
Carotid angioplasty and stenting (CAS) has emerged as an alternative treatment to endarterectomy in patients with carotid bifurcation lesions (1–3). Both approaches harbor the risk of embolic stroke (4–6). In CAS, the incidence of this potentially disastrous complication still needs to be established in a sufficiently large number of patients; yet, a variety of neuroprotective devices are currently being investigated as possible adjuncts to CAS to prevent embolic stroke. The first such device available to clinicians is the PercuSurge GuardWire, which can be used during interventions at degenerated saphenous vein grafts (7,8) as well as during carotid interventions (9). Its major component is a short balloon mounted on a hollow-tube guide wire by which the balloon can be placed and inflated distal to the target lesion. Blood flow across the lesion is thus blocked, and any atheromatous debris liberated during the intervention is prevented from entering the distal circulation. The present prospective study was designed to assess the device-specific advantages and disadvantages of GuardWire placement in the cervical vasculature and its short-term efficacy in preventing neurologic complications during CAS in a single-center setting.
Between January 1999 and March 2001, 96 consecutive patients (74 males; 68 ± 9 years) underwent elective CAS at our institution. Use of the GuardWire device was intended in all patients. Major patient inclusion criteria were the presence of a carotid bifurcation lesion with an angiographic diameter stenosis ≥70% (visual estimate) in symptomatic patients and ≥80% in asymptomatic patients, a distal diameter of the target vessel between 4 and 6 mm, and 40 mm of normal vessel distal to the lesion. Major exclusion criteria were total occlusion of the contralateral internal carotid artery (ICA) in the presence of ipsilateral vertebral or intracranial disease and angiographic evidence of thrombus at the lesion site, as well as a recent (≤14 days) stroke or myocardial infarction. All patients were informed of the investigational nature of CAS and the intended use of the GuardWire system and gave written, informed consent. Pertinent patient characteristics are summarized in Table 1.
The PercuSurge GuardWire temporary occlusion and aspiration system (Medtronic AVE, Santa Rosa, California) has been described in detail previously (7–9). In short, the three-component system consists of: 1) a 190- or 300-cm hollow guide wire with an outer diameter of 0.014 in. or 0.018 in. to which a compliant distal balloon is attached; 2) a MicroSeal adapter to enable balloon inflation, “sealing” of the wire to maintain balloon pressure after disconnection of the adapter, and balloon deflation; and 3) the Export aspiration catheter with a 1.1-mm distal lumen diameter.
An independent neurologist established the indication for intervention. The neurologic examination included a calculation of the National Institutes of Health (NIH) Stroke Scale and was repeated after the intervention and before discharge. Patients were discharged with a regimen of clopidogrel (75 mg/day for 1 month) and aspirin (100 mg/day).
Most patients received clopidogrel (75 mg/day) and aspirin (100 mg/day) at least three days before the intervention. In patients who were not premedicated, a loading dose of clopidogrel (300 mg) and a bolus of intravenous aspirin (500 mg) were administered before the intervention. A 100-cm 5F selective “Vitek” catheter (Cook Inc., Bloomington, Indiana) was used for brachiocephalic angiography. A 0.035-in. guide wire was then introduced through the lumen of the Vitek catheter into the external carotid artery, and the Vitek catheter was subsequently exchanged for a 6F or 7F long sheath. This was placed in the common carotid artery, and 70 to 100 IU/kg of heparin was administered to raise the activated clotting time above 225 s. An angiogram was obtained in the anteroposterior and lateral views to document cerebral blood flow. After advancement of the introducer sheath toward the carotid bifurcation, a magnified “worst view” angiogram of the lesion was taken.
After testing for integrity, the GuardWire balloon was fully deflated, introduced into the carotid artery, and advanced across the lesion. It was positioned at least 4 cm distal to the stenosis and slowly inflated until the vessel size was reached. Contrast dye was administered to verify total blockage of anterograde flow. At this point, atropine (1 mg) was given to prevent hypotension and bradycardia. A self-expanding Carotid Wallstent (Boston Scientific, Maple Grove, Minnesota) was then advanced into the lesion, and its correct position was checked before release by another administration of contrast dye, which ran off by way of the external carotid artery. Lesions were stented either directly or, in particularly high-grade stenoses, after predilation. All stents were postdilated with a 5.0-mm or 6.0-mm balloon, depending on the vessel size. After postdilation, the Export catheter was advanced toward the GuardWire balloon. Applying suction, the catheter was advanced and retracted once or twice between the stent and balloon until 20 to 40 ml of blood and debris were aspirated. Flushing the lesion site with saline solution was not performed. Finally, the distal occlusion balloon was deflated, and angiographic assessments were performed.
End points were evaluated on an intent-to-treat basis. Feasibility end points were the successful placement of the balloon distal to the lesion, device performance, and patient tolerance of balloon occlusion. The efficacy end point was any periprocedural neurologic event, including a transient ischemic attack (TIA), defined as a neurologic deficit that resolved without residuals within 24 h; a minor stroke, defined as a new neurologic deficit that resulted in slight impairment of neurologic function (speech, motor, or sensory skills) and either completely resolved within seven days or caused an increase in the NIH stroke scale of <4; and a major stroke, defined as a new neurologic deficit that persisted after seven days and increased the NIH stroke scale by ≥4 (10).
Exact 95% confidence intervals (CIs) based on the binomial distribution were calculated for proportions (11).
Ninety-six patients underwent a total of 102 interventions; six patients were treated sequentially for bilateral carotid disease. There were 54 culprit lesions located at the right internal carotid bifurcation and 48 at the left internal carotid bifurcation. The median estimated lesion length was 10 mm (range 4 to 30 mm), and the median estimated angiographic diameter stenosis was 90% (range 70% to 99%). Forty-seven lesions in 44 patients (46%) were associated with a history of neurologic symptoms within the previous six months (TIA or stroke), whereas 55 lesions in 52 patients (54%) had been asymptomatic. Figure 1 shows a breakdown of the 102 procedures and associated neurologic events.
The GuardWire balloon was successfully deployed in 99 procedures performed in 93 patients (97%). In two patients, the common carotid artery could not be accessed, because of an acute offset of the vessel. Both patients were referred to carotid endarterectomy. In the third patient who had a distal 360° loop of the left ICA, the GuardWire passed both the lesion and the loop but could not be advanced cranially enough to safely inflate the balloon (Fig. 2). The procedure was eventually completed without neuroprotection, using a standard 0.018-in. guide wire in exchange for the GuardWire.
Stenting was achieved successfully (residual stenosis <30%) in all 100 procedures completed on a catheter basis in 94 patients. Thus, the stenting failure rate was 2.1% (95% CI 0.2% to 7.3%). Direct stenting had been performed in 29 procedures. The median procedure duration and balloon occlusion times were 60 min (range 30 to 125 min) and 8 min (range 3.2 to 22.0 min), respectively.
Due to leakage of the GuardWire’s valve-sealing mechanism, balloon pressure was not maintained for the duration of the intervention in three patients (3.1%; 95% CI 0.7% to 8.9%). These patients must therefore be considered as having been treated without neuroprotection.
In a single patient, a dissection was noted after balloon deflation at the site of the balloon. This dissection was covered with another stent. There were no cases of stent malposition or bail-out stenting.
An adverse neurologic response to occlusion of the ICA that resolved promptly upon balloon deflation (balloon “intolerance”) was encountered in 5 (5.4%) of 93 patients (95% CI 1.8% to 12.1%). In three patients, symptoms (clouding of consciousness in 2 and mild paresis of the contralateral arm in 1) occurred 4, 5, and 6 min after balloon inflation and were tolerated until completion of the procedure after 7, 7, and 9 min, respectively. One patient, however, went into a coughing paroxysm and seizure after postdilation (4.5 min after balloon inflation), necessitating immediate deflation of the balloon before aspiration. In another patient, balloon inflation caused instantaneous loss of consciousness; stenting in this patient was subsequently performed without neuroprotection. These last two procedures had to be regarded as unprotected. They represent 2.2% (95% CI 0.3% to 7.6%) of patients in whom an adverse neurologic response to balloon occlusion of the target ICA prevented effective use of the GuardWire device.
Thus, with a total of eight procedures in which the GuardWire either could not be deployed (n = 3), did not maintain balloon pressure (n = 3), or was not tolerated to the extent that neuroprotected CAS was achieved (n = 2), 94 procedures were completed as intended in 88 patients, for a procedural feasibility rate of 92% (95% CI 84% to 96%).
On an intent-to-treat basis, periprocedural neurologic complications were observed in five patients (5.2%; 95% CI 1.7% to 11.7%). There were three strokes (3.1%) and two TIAs. Two minor strokes occurred in symptomatic patients, whereas a major stroke and both TIAs were observed in asymptomatic patients, for major complication (stroke) rates in symptomatic and asymptomatic patients of 4.6% (95% CI 0.6% to 15.5%) and 1.9% (95% CI 0.1% to 10.3%), respectively. Both minor strokes occurred in patients in whom the device could not be deployed; the major stroke and TIAs were encountered in patients in whom the device had been in place throughout the intervention (Table 2). Thus, balloon-protected CAS could be performed according to the protocol without major neurologic events in 87 patients, for a procedural success rate of 91% (95% CI 83% to 96%). The periprocedural rate of stroke/TIA, even though the device was in place, was 3.4% (3 of 88 patients; 95% CI 0.7% to 9.6%).
The patient who sustained a major stroke was a 58-year-old man with severe peripheral artery occlusive disease who experienced contralateral hemiparesis immediately after the intervention at an asymptomatic high-grade stenosis of the right ICA; five weeks after the intervention, left-sided hypoesthesia and impaired coordination of the left hand and leg were still present.
Of the minor strokes, one occurred in the patient with a 360° ICA loop (Fig. 2). Attempts to place the GuardWire appropriately distal to this anatomic abnormality lasted for about 20 to 25 min and were eventually aborted in favor of unprotected stenting. Neurologic symptoms related to the ipsilateral hemisphere were noted within minutes of completion of the procedure; they comprised weakness of the right arm and discrete aphasia. Symptoms resolved completely within a few days.
In another (female) patient, about 90 min was spent in ultimately futile attempts to cannulate the right common carotid artery, using both transfemoral and transbrachial routes. Three hours after termination of the procedure without any therapeutic intervention, the patient developed a minor stroke with symptoms (hemiparesis, mild aphasia) related to the contralateral hemisphere; symptoms resolved within 48 h.
This single-center, prospective study of 96 patients showed that placement of the PercuSurge GuardWire distal to a high-grade carotid bifurcation lesion can be achieved in 97% of patients. However, failure of the balloon to maintain pressure throughout the intervention (n = 3) and severe neurologic responses to balloon occlusion of the targeted ICA (n = 2) reduced the procedural feasibility rate to 92% in our patient cohort. There were no deaths. A total of five periprocedural neurologic complications (5.4%) were observed, three of which were strokes (3.1%). The strokes occurred in two symptomatic patients and one asymptomatic patient, resulting in major periprocedural complication rates in these patient subgroups of 4.6% and 1.9%, respectively. Two strokes were encountered in patients in whom the device could not be deployed, whereas the other stroke and two TIAs occurred despite per-protocol GuardWire placement and function. Thus, procedural success (device placement and stenting without major neurologic events) was achieved in 87 patients (91%).
Adverse vessel anatomy prohibited introduction of the device into the common carotid artery in two patients, one of whom sustained a minor contralateral stroke. In another patient with a distal ICA loop, the GuardWire could not be placed distally enough to safely inflate the balloon, and stenting was performed without neuroprotection; the patient had a minor ipsilateral stroke. Neither event appears to be device-specific.
Once the GuardWire balloon is successfully positioned and inflated, stenting equipment will be advanced over the wire. Despite blockage of anterograde ICA flow, correct positioning of the unexpanded stent is still possible through the injection of a contrast agent, which will drain into the external carotid artery. However, friction between the balloon and vessel wall during the manipulation of endoluminal equipment may cause a dissection, as observed in one of our patients.
Leakage of the GuardWire’s valve-sealing mechanism resulted in a gradual loss of balloon pressure during three procedures; consequently, these were essentially concluded without neuroprotection. The underlying technical problem has been recognized by the manufacturer and remedied in the currently available GuardWire Plus system.
A total of five patients experienced an adverse neurologic response to balloon occlusion that subsided without residuals upon balloon deflation. In four patients, symptoms occurred within minutes of starting the stenting procedure; these symptoms could be controlled until its conclusion in three patients. In the fourth patient, however, the symptoms were so severe that immediate balloon deflation was necessary before aspiration, rendering the procedure unprotected. The final patient lost consciousness instantaneously upon balloon inflation, prohibiting further use of the device. Thus, balloon intolerance is a system-immanent disadvantage that may prohibit system use or render it ineffective in about 2% of cases. Conclusive factors predictive of balloon intolerance have not yet been established. With the present availability of filter systems for neuroprotection during CAS (12), interventionists may convert to such a device in cases of balloon intolerance.
Major neurologic events related to the intervention were observed in three study patients (3.1%). This incidence is lower, yet not statistically significantly so, than that reported for CAS procedures performed without neuroprotection (1–3). Yadav et al. (1) encountered eight periprocedural strokes and one death in 107 consecutive patients treated between March 1994 and November 1995, for a major procedure-related complication rate of 7.5%. In a global survey of 4,757 patients investigated by the end of September 1999 at 36 centers, 200 strokes and 41 procedure-related deaths were reported, for a combined stroke/death rate of 5.1% (2). Roubin et al. (3) recently published an update of their previous study (1); they encountered 38 nonfatal strokes (7.2%) and five deaths (1.0%) in 528 consecutive patients. The lack of a statistical significance of the difference in incidence between our study and those mentioned may be due to a type-2 statistical error (low number of patients). However, it may also indicate that, on an intent-to-treat basis, there may actually be no reduction in the major neurologic complication rate associated with balloon-protected versus unprotected CAS. This is apparently supported by our finding in two patients that “mere,” and ultimately futile, attempts at cannulating the target artery or placing the protective device distal to the lesion can result in major neurologic complications.
The American Heart Association guidelines for carotid endarterectomy recommend intervention in symptomatic patients with ipsilateral lesions ≥70% angiographic diameter stenosis, if the stroke morbidity and mortality rate can be kept below 6%, and in asymptomatic patients with ≥60% diameter reduction, if the interventional risk of major complications can be kept below 3% (13). The stroke rates observed for symptomatic and asymptomatic patients in this study (4.6% and 1.9%, respectively) were well within these limits.
This study was not randomized and therefore lacked a control group of patients. Quantitative angiography was not used to measure lesion length and diameter stenosis. No objective evaluation of embolic events by means of computed tomography or magnetic resonance imaging was performed. Histology of aspirated debris has not been routinely performed in this study; this aspect has been addressed in a previous report on a subgroup of patients (14).
It is concluded that use of the GuardWire temporary balloon-occlusive system as an adjunct to CAS is feasible in most patients, with a 5.4% overall incidence of periprocedural neurologic complications and a 3.1% major and minor stroke rate. Adverse neurologic reactions to balloon occlusion may prohibit effective use of the system in about 2% of patients. Superiority of balloon-protected CAS over previously reported unprotected CAS could not be demonstrated with statistical significance and may warrant a larger comparative trial.
☆ This study was supported by PercuSurge, Inc., Sunnyvale, California.
- carotid angioplasty and stenting
- confidence interval
- internal carotid artery
- National Institutes of Health
- transient ischemic attack
- Received December 27, 2001.
- Revision received April 25, 2002.
- Accepted May 24, 2002.
- American College of Cardiology Foundation
- Yadav J.S.,
- Roubin G.S.,
- Iyer S.,
- et al.
- Roubin G.S.,
- New G.,
- Iyer S.S.,
- et al.
- Alberts M.J.
- Endovascular versus surgical treatment in patients with carotid stenosis in the Carotid And Vertebral Artery Transluminal Angioplasty Study (CAVATAS): a randomised trial. Lancet 2001;357:1729–37
- Carlino M.,
- De Gregorio J.,
- Di Mario C.,
- et al.
- Baim D.S.,
- Wahr D.,
- George B.,
- et al.,
- the Saphenous vein graft Angioplasty Free of Emboli Randomized (SAFER) Trial Investigators
- Brott T.,
- Adams H.P. Jr.,
- Olinger C.P.,
- et al.
- Campbell M.J.,
- Machin D.
- Reimers B.,
- Corvaja N.,
- Moshiri S.,
- et al.
- Biller J.,
- Feinberg W.M.,
- Castaldo J.E.,
- et al.
- Tübler T.,
- Schlüter M.,
- Dirsch O.,
- et al.