For years, distal embolization following revascularization of degenerated, aged saphenous vein grafts has been a well-recognized complication. The search for effective treatment of vein graft atherosclerosis has been a paramount challenge for cardiovascular medicine. Experience with new percutaneous interventional devices and pharmaceutical agents has been disappointing. Likewise, application of percutaneous intervention to carotid artery stenosis has been limited by the neurological consequences of cerebral embolization. To address the problem, an exciting new approach is to use a mechanical device to prevent particulate matter from traversing distally. This article reviews the promises of novel distal protection devices to reduce the risk or ameliorate the sequelae of embolization during percutaneous revascularization.
Dr. Tan is a Fellow in Interventional Cardiology at the
University of Maryland Medical Center, Baltimore, MD.
A major limitation of coronary artery bypass graft surgery
(CABG) has been the progression of atheromatous disease in
aortocoronary saphenous vein grafts (SVGs). SVG attrition is
approximately 7% during the first week, 15% to 20% during the first
year, 1% to 2% per year from 1 to 6 years, and 4% per year from 6
to 10 years after surgery. At 10 years, only 40% of patent grafts
are free of significant stenosis.
Repeat bypass surgery has higher risk than the initial procedure.
Diseased SVG interventions are limited by distal embolization,
which may arise from disruption of soft, friable atherosclerotic
plaque and adherent thrombus. The reported incidence of distal
embolization following balloon angioplasty of SVGs ranges from 2%
resulting in no reflow, myocardial infarction (MI), and death.
These complications do not appear to be significantly reduced with
glycoprotein (GP) IIb/IIIa inhibition.
Carotid stenting is being evaluated as an alternative to surgery
for primary carotid occlusive disease. However, there is an
important risk of periprocedural stroke, undoubtedly related to
embolization. In carotid interventions, virtually every patient has
acoustic and Doppler signal evidence of cerebral embolization.
Using transcranial Doppler ultrasound, the number of detected
embolic particles during standard carotid endarterectomy correlated
with the rate of subsequent neurologic events.
In addition, renal artery stenting is currently an accepted
treatment option for atherosclerotic renal artery stenosis.
Atheroembolism during the procedure seems to be an important factor
for subsequent decline in renal function in a subset of
More recently, evidence from multiple fronts has underscored the
frequency and prognostic importance of embolization during and
acute coronary syndromes, which can be promoted by catheter-based
To overcome these problems, devices for containment of embolic
material have been developed. There are two general types of distal
protection devices: occlusive devices and filters. The occlusive
device has a balloon that occludes the artery during intervention
with aspiration of debris with a small catheter. Filters trap
debris during intervention and are then collapsed and withdrawn
from the artery with the trapped debris. Theoretically, occlusive
devices allow more complete emboli capture. However, end-organ
ischemia, vessel trauma, distal vessel thrombosis, and poor distal
vessel visualization are limitations of these devices. The filter
devices have the advantage of maintaining distal perfusion with the
disadvantage of larger crossing profiles and their inability to
capture the smallest particles. The practical lower limit for pore
size appears to be 50 µm. Smaller microparticulate matter can still
get through the filter, although particles that small may have no
Importantly, use of emboli protection devices adds procedural time
and complexity to a technique in which manipulation should be
The concept of a distal balloon occlusive device was first
described in 1987 by Theron et al
for cerebral protection during carotid angioplasty. In recent
years, this notion has gained popularity.
PercuSurge (Medtronic AVE, Santa Rosa, CA) is the device that
has undergone the most clinical testing, and in almost all cases,
embolic material was retrieved. This device has just been approved
for general clinical use in the United States.
The PercuSurge GuardWire temporary occlusion system (Medtronic
Inc.) provides temporary vascular occlusion during percutaneous
intervention. The PercuSurge System consists of 3 components
: 1) GuardWire; 2) the MicroSeal Adapter; and 3) the Export
aspiration catheter. The GuardWire is a 190-cm-long, 0.014-in
hollow guidewire with a central lumen connected to a compliant
distal occlusion balloon with a 0.41- to 0.43-in crossing profile
and available inflated diameters of 3.5 to 5 mm (figure 1). Regular
balloon catheters and stent delivery systems can be advanced over
this wire to perform percutaneous interventions. A proprietary
valve sealing system at the proximal end of the guidewire maintains
distal balloon inflation despite disconnection of the inflation
device. The MicroSeal Adapter is a device that controls the opening
and closure of the proximal valve, allowing inflation and deflation
of the distal balloon. The The Export aspiration catheter is a 5F
monorail catheter connected to a 20-mL syringe, providing a
low-pressure vacuum to remove debris and thrombus from the graft.
The 135-cm-long aspiration catheter has a 35-cm-long distal
monorail wire lumen. A schematic representation of the PercuSurge
System is shown in figure 2.
The initial clinical experience with PercuSurge was limited to SVG
interventions; however, the principles of its application can be
extended to most vascular structures. It is particularly suited to
vessels without major side branches, which helps to ensure that all
emboli are contained by the inflated GuardWire.
Saphenous vein graft interventions
In 1999, Webb et al
reported the results of a pilot study of 22 patients who underwent
27 SVG stenting procedures using the PercuSurge system. Distal
graft occlusion time averaged 150 ± 54 seconds and decreased as
experience was gained. The procedure was well tolerated. The
in-hospital event rates, including creatine kinase (CK) elevation
(11.1%) and non-Q wave MI (3.7%), were lower than those in historic
controls. Particulate material was retrieved in 91% of cases.
Particle size was 204 ± 57 µm in the major axis and 83 ± 22 µm in
the minor axis. Retrieved particulates consisted predominantly of
soft acellular atheromatous material, such as is typically found
under a fibrous cap. In no case was vessel wall damage seen as a
result of the distal occlusive balloon, and late development of a
stenosis at the distal occlusion was not apparent. Another
PercuSurge registry, the SAFE trial
reported similarly low rates of in-hospital major adverse clinical
events (MACEs) for SVG interventions. A total of 103 patients were
enrolled in this multicenter registry and the in-hospital MACE rate
was 4.9%. Visible material was removed in 95% of cases, and 81% of
the material was less than 96 µm.
The Saphenous Vein Graft Angioplasty Free of Emboli Randomized
(Transcatheter Cardiovascular Therapeutics meeting, 2000) was the
first prospective multicenter randomized trial to determine if
PercuSurge reduced the incidence of 30-day MACEs compared with
unprotected stenting of SVGs. The study included lesions in SVGs 3
to 6 mm in diameter, more than 5 mm from the ostium and more than
20 mm from the distal anastomosis. Patients with ongoing myocardial
infarction, ejection fraction <25%, serum creatinine >2.5,
and planned use of atherectomy were excluded. A total of 801
patients were enrolled. Use of the PercuSurge system was
technically successful in 91.6% of the patients. Procedural success
was achieved in 90.7% of the patients in the PercuSurge group and
in 84% of the control group.
In the PercuSurge group, there was a 49% reduction in the 30-day
primary composite end point of MI, death, emergent bypass surgery,
and target lesion revascularization (TLR) (9.0% vs. 17.8%,
= 0.001). Most of the difference was due to a significantly lower
rate of non-Q wave MI in the PercuSurge group (6.9% vs. 14.5%,
= 0.003). In addition, there was a lower incidence of no-reflow in
the PercuSurge group (3.3% vs. 8.3%,
= 0.005) with no significant difference in the incidence of
perforation, dissection, or subacute closure. The Percu-Surge group
had a lower incidence of postprocedure TIMI grade 2 flow (0.9% vs.
= 0.004), whereas the incidence of TIMI grade 0, 1, and 3 flow were
similar among the two groups. GP IIb/IIIa inhibitors were used in
>60% of cases in both arms. The MACE benefit of PercuSurge was
independent of GP IIb/IIIa inhibitor use.
Evidently, the SAFER trial is a landmark trial that demonstrates
the safety and efficacy of PercuSurge in recovering embolic debris,
preserving normal flow, and reducing MACEs during SVG
Using transcranial Doppler, Al-Mubarak et al
studied the frequency of microembolic signals during carotid
stenting. With PercuSurge protection, the frequency of the
Doppler-detected microembolic signals was significantly reduced
compared with the frequency of the signals with no protective
device. Microembolic signals in the PercuSurge group occurred
predominantly during sheath placement and wiring and during the
occlusion balloon deflation.
Small studies of carotid interventions with PercuSurge have
shown that cerebral protection with PercuSurge is feasible, safe
and effective. Henry et al
reported their single-center experience of 148 patients who
underwent 164 internal carotid artery stenting procedures with
PercuSurge protection. Immediate technical success was achieved in
99.4% of cases. Mean flow occlusion time was 422 seconds (range 125
to 1479 seconds) and the procedure was well tolerated in 95.1% of
cases. Debris was retrieved in all patients. The in-hospital
neurologic complication rate was 1.8%. The Carotid Angioplasty Free
of Emboli (CAFE-USA) trial
reported similarly favorable results. This phase 1 multicenter
registry evaluated the use of PercuSurge in 70 patients who
underwent carotid stenting. Procedural success was 100%, and
prolonged distal occlusion was well tolerated (mean time 12.5
minutes, range 3.4 to 35.6 minutes). Macroscopic debris was
aspirated in all patients. The 30-day complication rate (stroke,
transient ischemic attack [TIA], death) was 8.5%.
However, another single-center study
compared outcomes of carotid stenting with and without PercuSurge
versus carotid endarterectomy and found disappointing results with
carotid stenting despite protection. There were significantly more
in-hospital complications (death, TIA, stroke) in all stent
patients than with the surgical cohort patients (22.2% vs. 3%,
= 0.0077). Although PercuSurge decreased the in-hospital
complication rate (15% vs. 27% for unprotected stenting), the
outcomes were overwhelmingly worse than in the surgical group.
Without question, the efficacy of PercuSurge in preventing
neurological complication will require a large randomized trial
comparing protected carotid stenting with surgical
Percutaneous intervention in AMI
A single-center series of 20 patients who underwent protected
percutaneous intervention for AMI with PercuSurge in native
coronary arteries suggested that this approach is feasible in
Case selection consisted of patients with large thrombi in large
vessels determined angiographically (proximal reference diameter
3.6 ± 0.7 mm), including: 7 left anterior descending, 11 right
coronary, and 2 left circumflex. The procedure was successful in 18
of 20 patients. Macroscopically visible material was obtained in
all successful cases, with histology demonstrating fresh and/or
partial organizing thrombus. Angiographically evident embolization
occurred in only 1 patient. Others reported improved angiographic,
epicardial, and myocardial perfusion in AMI patients who underwent
stenting of the infarct-related artery with PercuSurge protection.
Larger trials of primary angioplasty with distal protection are in
Renal artery interventions--
Henry et al
reported a pilot study of 26 patients who underwent protected renal
angioplasty and stenting with the PercuSurge device. Immediate
technical success was achieved in all cases. Visible debris was
extracted from all patients. At 6 months, there was no renal
function deterioration in any patient. These preliminary data
suggest that protected renal artery revascularization is feasible
Parodi Anti-Emboli System
Parodi et al
have developed a circulatory control device system, the Parodi
Anti-Emboli System (PAES), which creates temporary blood flow
reversal from the carotid artery into a suctioning catheter
positioned proximal to the carotid artery target lesion. This
device, specifically designed for the carotid system, is quite
complicated and bulky.
The PAES device consists of two components: a 7F or 8F guiding
catheter with an inverted pear-shaped balloon at the tip for
occlusion of the common carotid artery through which the
low-profile, guidewire-based, balloon-type device is used for
occlusion of the external carotid artery. The guiding catheter
lumen is then connected externally to the contralateral femoral
vein introducer with an interposed blood filter, which achieves
inversion of the internal carotid blood flow. During angioplasty
and stenting, additional suction is used to capture embolic
A multicenter registry of 60 patients using the PAES during carotid
stenting showed no in-hospital strokes, TIA, loss of consciousness,
There were 3 patients who were intolerant to flow reversal without
The concept of a filter device was first introduced in 1991 by
Gunther and Vorwerk
for use in peripheral arterial intervention. Recently, several
other filters have been designed and are in various stages of
testing, although they have not been approved for general clinical
use in the United States. This section will review the filter
devices that have preliminary experience in humans.
The AngioGuard Guidewire system (Cordis J&J, Warren, NJ) has
undergone the most clinical evaluation of the filter device
The AngioGuard Guidewire system uses a distal guidewire-based
filter that traps debris. The filter device is a flexible
olive-shaped Nitinol structure, distally covered by a polyurethane
filter connected to a long, 0.014-in wire. The closed device's
crossing profile is 3.4F and is available in 50-µm and 100-µm
pores. After crossing the target lesion, the basket (4 to 7 mm in
diameter) is expanded by removal of the delivery sheath, and the
intervention can be performed in a standard manner. Plaque debris
is captured within the distal basket, and at the end of the
procedure, the filter is closed and retrieved. The filter membrane
allows normal blood flow during the procedure.
In a small number of patients, AngioGuard has been tested in
percutaneous coronary, SVGs, and carotid and renal interventions,
with a resulting retrieval of atherosclerotic material in almost
all of those patients.
Two randomized trials are in progress to evaluate the clinical
impact of the AngioGuard device. The SAPPHIRE trial is a randomized
trial of protected carotid stenting versus surgical carotid
endarterectomy. The GUARD trial compares SVG interventions
randomized to be performed with or without AngioGuard.
TRAP Vascular Filtration System
The TRAP Vascular Filtration System (VFS) (Microvena, White Bear
Lake, MN) consists of a specially treated nitinol braided basket
that captures embolic material during percutaneous intervention.
The basket diameters are available from 2.5 to 7 mm, in 0.5-mm
In Europe, 32 patients with de novo SVG lesions were treated
successfully with the TRAP VFS.
Debris was captured in all but one basket. There were no
device-related MACEs at 30 days.
EPI FilterWireEX device
EPI FilterWireEX device (Boston Scientific/EP Technologies, San
Jose, CA) is a temporary intra-arterial filtration system using a
standard 0.014-in guidewire that incorporates a round nitinol wire
loop structure that supports a thin, 75-µm porous filter membrane
at the distal end. The filter is made of polyethylene and rotates
freely on the end of the guidewire.
A small number of patients have been treated for SVG, native
coronary, and carotid stenosis with the EPI FilterWireEX device.
No in-hospital complications were reported.
Embolic particulate matter is commonly induced by percutaneous
revascularization. This particulate matter may play a pivotal role
in the pathogenesis of distal embolization and microvascular
obstruction following percutaneous intervention. Use of innovative
emboli protection devices to shield the microvasculature may
improve short- and long-term outcome. It is still early in the
development of these novel devices. Making them simple and easy to
use and as atraumatic as possible, with the lowest profile and
highest torqueability, will require further refinements. Randomized
controlled trials for each distal protection device and each
vascular bed are warranted in order to prove that interventions are
accomplished more safely. Undoubtedly, a safe and effective device
will become available and, ultimately, will be incorporated into
the daily practice of percutaneous coronary and peripheral