ABSTRACT: Background. Prevention of distal embolization during percutaneous coronary revascularization may be necessary to reduce post interventional morbidity and mortality.

Methods. We employed a newly developed emboli containment and retrieval system in native coronary arteries during percutaneous coronary angioplasty and stenting in 39 selected patients (mean age, 58.9 ± 10.1 years; 11 females) presenting with acute [n = 22; 8 left anterior descending coronary arteries (LAD), 3 left circumflex arteries (LCX), 11 right coronary arteries (RCA)], subacute (n = 7; 2 LAD, 2 LCX, 3 RCA) or chronic (n = 6; 2 LAD, 4 RCA) total or subtotal occlusion of an infarct-related vessel, or with severe stenosis and symptoms of unstable angina (n = 4; 2 LAD, 2 RCA). Protection device-assisted angioplasty with stent implantation was uneventful in all patients with good angiographic results and normal post procedural flow. Intermittent aggravation of anginal pain during inflation of the occlusive balloon (from 2.5 to a maximum of 25 minutes cumulative inflation time) was observed in 19 of the 36 conscious patients (7 with acute, 7 with subacute and 3 with chronic occlusion, and 2 with unstable angina), but caused neither interruption of distal occlusion nor hemodynamic instability. In 31 patients, the aspirates contained visible debris. Histological analysis showed particles up to 12 mm in size, consisting of necrotic core, inflammatory cells, cholesterol debris, and old and fresh thrombi. In 8 patients, the aspirated particles were too small to allow microscopic diagnosis or debris was absent.

Conclusions. This preliminary report demonstrates the feasibility of using a protection device in native coronary arteries to prevent distal embolization of particulate matter that is mobilized during percutaneous interventions. To the extent that this material contributes to the mechanisms of distal embolization, no-reflow and infarction, this device may help to reduce such complications. Appropriately designed trials are required to assess the clinical benefit of this system.

Recent developments to improve percutaneous reperfusion strategies include transluminal extraction atherectomy,¹ excimer laser coronary angioplasty (ELCA),² angiojet thrombectomy,³ coronary ultrasound thrombolysis (CUT)⁴ and, among others, aspiration thrombectomy (ICAT).⁵ In an effort to reduce complications such as distal embolization and no-reflow, a new catheter system (PercuSurge GuardWire™ Temporary Occlusion and Aspiration System; PercuSurge, Inc.) was designed⁶ and first evaluated in patients during saphenous vein graft (SVG) interventions.^7,8 Both reports^7,8 have convincingly demonstrated the usefulness against distal embolization during percutaneous interventions. However, the risk of peripheral embolization, no-reflow and myocardial infarction caused by mobilization of particles from obstructive plaques, organized thrombi and/or newly formed atherothrombotic material is not limited to venous grafts; native coronary arteries might be exposed to a similar risk during conventional angioplasty or stenting and these complications might severely affect morbidity and the clinical outcome in these patients as well.

The promising results obtained in saphenous vein graft procedures prompted us to evaluate the feasibility and usefulness of this new device in native coronary arteries in preventing peripheral embolization of friable material during coronary interventions.

Methods
We report our initial experience in the first series of patients that underwent PTCA and stenting utilizing the first generation PercuSurge GuardWire™ protection device in native coronary arteries.

Device description and procedures. In brief, and as described elsewhere,^6–8 the first generation catheter system consists of a 0.014-inch guidewire with a compliant balloon (two sizes: 3.0–4.0 and 4.0–5.0 mm, respectively) mounted 32 mm proximally of the soft tip (Figure 1). It is supplied with a radiopaque marker to facilitate proper positioning after crossing of the lesion. A detachable inflation adapter activates the special valve sealing system of the wire for inflation (< 2 atm) and deflation with radiopaque contrast to ascertain adequate temporary occlusion of the vessel (range, 3–6 mm) during interventions. Regular angioplasty balloon or stent delivery balloon catheters as well as the aspiration catheter are easily exchanged and advanced over this wire. Retrieval of potentially embolic material is started during or is performed at the end of PTCA/stenting by means of a monorail catheter (Export™ Aspiration catheter, maximum thickness 5.4 Fr) that is fully advanced to the temporary occlusion balloon for repeated vacuum aspirations using 20 ml syringes. Syringes are emptied and the content of the aspiration catheter is rinsed over a cell strainer (40 mm nylon) serving as a filter. The solid material is semiquantitatively counted and analyzed for size and color and then immersed in formalin (4%) for histologic examination.

Patient selection. We selected 39 suitable patients (mean age, 58.9 ± 10.1 years, 11 females) for this pilot protocol. Informed consent was obtained from all conscious patients or from their relatives in unconscious patients. Patients were eligible when presenting without severe hemodynamic instability or relevant rhythm disturbances. In all but one patient, the procedure was performed immediately after diagnostic coronary angiography. Ongoing myocardial infarction, old occlusion, long lesions or pretreatment strategies (including failed thrombolysis in 3 patients) were not exclusion criteria since we were primarily interested to test the feasibility and the utility of the device in native coronary arteries. Left ventricular function was assessed either by angiography (n = 21) or by echocardiography (n = 12) obtained within 3 days of the intervention.

Patients were included when the following angiographic requirements were met: the guiding catheter (8 Fr) was easy to place, the culprit vessel had a suitable size (>= 3.0 mm) and the lesion was located in the proximal or midportion of the vessel. Patients with main branchings in proximity to the target and relevant tortuosities were regarded as less suitable candidates for this procedure.

Patients were categorized as having acute myocardial infarction (onset of symptoms up to 12 hours), subacute myocardial infarction (< 12 hours to 4 days), or chronic myocardial infarction (> 4 days). Patients with unstable angina had persistent anginal episodes despite maximal antianginal therapy and severe stenosis of the culprit lesion, but no evidence of myocardial necrosis.

Peri- and postinterventional management. In order to assess the feasibility of the device, we primarily attempted to cross the culprit lesion with the PercuSurge GuardWire™ first. However, there was no restriction to pre-wire the vessel in case the operator decided to do so. In some instances, this was judged to be necessary to ascertain safe placement of the wire distal to the lesion and/or to path a channel for the protection device. Similarly, the decision to predilate a vessel before placing the GuardWire™ or predilate the vessel before stenting was made by the operator based on his experience and his judgement how to perform the safest intervention and to produce the best procedural outcome. Once in place, the position of the GuardWire™ was checked by fluoroscopy and by injection of contrast dye, and the position of the device was adjusted if needed to allow regular PTCA- or stent-balloons to reach the lesion; on the other hand, it was important not to push the GuardWire™ too far distal from the lesion because of the risk of leaving sidebranches unprotected. The distal balloon was then inflated under fluoroscopic control, and contrast media was injected to test completeness of the distal occlusion. All interventions were performed over this wire with the distal balloon inflated; importantly, no more adjustments of the GuardWire™ were done once the PTCA/stenting procedures had started.

Following the last aspiration, the distal balloon was deflated. The result was visualized; if additional interventions were needed, they were all performed under distal protection again. The sum of each single distal occlusion time was defined as total occlusion time. Finally, the GuardWire™ was pulled back. Thereafter, a final angiographic control was performed which permitted study of the intervention site, as well as that of occlusion of the device’s balloon in order to assess any damage to the vessel. Moreover, particular attention was paid to the presence of distal embolization manifesting itself as interruption of radiopaque contour or abrupt closure, or generalized diminution of flow (slow- or no-reflow).

Glycoprotein IIb/IIIa inhibitors were given according to the decision of the operator. A bolus of heparin (150–200 U/kg body weight) was administered prior to intervention in order to achieve an initial activated clotting time (ACT) of > 300 seconds or of 200–250 seconds in the presence of glycoprotein inhibitors, respectively. A bolus of 500 mg of aspirin i.v. was given immediately after stent implantation.

Post-procedural management consisted of clopidogrel 300 mg on the day of intervention and 75 mg daily thereafter for one month and concomitant aspirin 100 mg daily as continuous therapy. The sheath was removed after the procedure and the puncture site was closed using the Perclose™ system in all patients.

Definitions. The culprit lesion was determined by its anatomical location, its perfusion characteristics according to Thrombolysis in Myocardial Infarction (TIMI) classification for flow through the infarct-related vessel (Table 1)⁹ and for collateral flow to the infarct-related vessel (Table 2),¹⁰ morphological features and intraluminal thrombus. Intraluminal masses were described as thrombus (apparent or possible thrombus) when contours and mural opacities were rounded or polypoid in shape, and protruded into the lumen. In totally occluded vessels, the abrupt termination with a stump and delayed wash-out of contrast were regarded as recent thrombotic occlusion. Smooth tapering up to terminal sidebranches before sudden interruption of the main vessel and in general, well-developed collateralization in the distal portion was defined as chronic occlusion. When visualized, the severity and length of the obstruction were analyzed qualitatively by comparing the contours of the greatest narrowing to the densely opacified borders before and after the lesion.

Statistics. Data are reported as means ± SD. Comparison between groups was performed using the unpaired Student’s t-test (Statview IV; Abacus Concept, Inc., Berkeley, California). A level of p < 0.05 was considered to be significant.

Results
Clinical and angiographic findings of the patients are listed in Table 3. Patients presented with acute (n = 22; 8 LAD, 3 LCX, 11 RCA; mean time from onset of symptoms to intervention: 5.0 ± 2.9 hours), subacute (n = 7; 2 LAD, 2 LCX, 3 RCA) or chronic (n = 6; 2 LAD, 4 RCA) occlusion or subtotal occlusion (mean diameter of stenosis: 98.9 ± 2.2%; range: 90–100%) of an infarct-related vessel, or had unstable angina (n = 4; 2 LAD, 2 RCA; mean diameter of stenosis: 94.8 ± 3.7%; range: 90–99%). In 25 patients, coronary angiography clearly showed presence of thrombotic material in proximity of the lesion (Figure 2). Left ventricular ejection fraction on average was mildly impaired (56.4 ± 13.0%; range: 30–81%).

Table 4 summarizes interventional procedures, adjunctive management and procedural outcome. In 22 interventions, the lesion was crossed with the GuardWire™ first without prewiring or predilation, in 4 cases predilation was required. In 1 patient, two protection devices were simultaneously used because his LAD lesion was very proximal to the origin of a large diagonal branch (Figure 3); this procedure was uneventful as well as in all other patients. Intention to cross the lesion first with the GuardWire™ failed in 1 patient only. Direct stenting was performed in 28 patients; in 11 patients, balloon predilation was required followed by stenting.

Total occlusion time ranged from 2.5 to 25 minutes (mean time, 9.7 ± 5.9 minutes); it did not differ in patients with chronic or subacute (n = 16; 11.3 ± 7.1 minutes; range: 3.5–25 minutes) as compared to patients with acute infarction or unstable angina (n = 23; 8.6 ± 4.8 minutes; range: 2.5–23 minutes). Direct stenting (n = 28; 8.0 ± 4.4 minutes) markedly reduced occlusion time as compared to stenting after predilation (n = 11; 14.9 ± 6.1 minutes; p < 0.001). Angiographic presence of thrombus prolonged the intervention (n = 25; 11.0 ± 6.0 minutes), but this was not significant compared to patients without thrombus (n = 14; 7.7 ± 5.2 minutes; p = NS). Repeated and long occlusion times were mostly required in vessels with long thrombotic lesions. The average number of aspirations performed was 4 ± 2.6 ml (range: 2–15 ml) and 65.0 ± 37.7 ml (range: 30–235 ml) of blood was aspirated.

Vessel occlusion with the device’s balloon was associated with a reversible increase of anginal symptoms in 19 patients (7 acute, 7 subacute, 3 chronic infarction, and 2 unstable angina), but worsening of angina never required premature termination of the inflation or interruption of the procedure. Symptoms, however, could not be assessed in 3 patients because of mechanical ventilation after cardiopulmonary resuscitation for complications of acute myocardial infarction. Similar to angina, hemodynamic instability during temporary occlusion that might have led to discontinuation of the inflation of the device’s balloon never occurred.

All patients had TIMI grade III flow at the end of the intervention except 4 patients with TIMI grade II-III flow. Review of the final angiograms showed no damage to the vessel wall at the site of the inflation of the device’s balloon.

In 5 patients with subtotal occlusions, abrupt discontinuation of the radiopaque contour in the periphery of an infarct-related vessel was detected prior to the intervention, suggesting peripheral embolization. In 2 patients, new embolizations were seen in small sidebranches following the intervention suggesting that minor embolization during the intervention could not be fully prevented by the device. In 24 patients (62%), a glycoprotein IIb/IIIa inhibitor was given either before or during (n = 21), or at the end of the intervention (n = 3), respectively.

Table 5 lists the semiquantitative macroscopic appearance of the aspirates and the microscopic/histologic diagnosis of the material. In 31 patients, the aspirates contained visible debris (>= 0.5 mm; n = 27) or small particles (< 0.5 mm; n = 4) (Figure 4A). Histological analysis showed particles up to 12 mm in size (Figure 4B) consisting of older and fresh thrombi (Figure 5), and/or particles of atherosclerotic plaques with foam cells, lipid deposits and cholesterol crystals (Figure 6). In 9 patients, the aspirated particles were either too small to allow microscopic diagnosis or no debris was found. Interestingly, in 8 patients in whom coronary angiography did not suggest intracoronary thrombus, both macroscopic as well as microscopic examination revealed presence of thrombus. Conversely, in 3 patients, there was no macroscopic or microscopic evidence of thrombus despite clearly visible thrombus-like opacifications.

Follow-up. All 39 patients underwent the protection device-assisted percutaneous intervention without any procedure-related complications. Follow-up visits or telephone interviews (range: 12–134 days) yielded an uneventful course after hospital discharge in 88% of patients. One patient died 7 days after recanalization of the RCA because of anoxic encephalopathy during sustained cardiopulmonary resuscitation before hospitalization. Within 2 days, two patients underwent a second intervention in the same vessel because of acute reocclusion with successful recanalization (both occurring in the RCA).

Discussion
This pilot study shows that during angioplasty and stenting of totally occluded or subtotally occluded coronary arteries in patients with myocardial infarction and unstable angina, there is mobilization of friable material which can be retrieved and thus be prevented from peripheral embolization in the vast majority of patients. This was achieved without any obvious damage to the vessel at the inflation site of the device’s balloon, attesting to the safety of this system.

In this pilot trial, we selected patients based on the operators’ analysis of the coronary angiogram and their assumption that the coronary anatomy, particularly the size of the vessel (>= 3.0 mm) and the type of lesion, was suitable for this procedure. Accordingly, crossing of the lesion with the device was easy in most instances except in one patient where the intention to pass the lesion with the GuardWire™ first failed. In some instances pre-wiring was helpful, i.e., when the guiding catheter did not offer sufficient back-up.

To date, there is no solid proof whether patients do benefit from protection device-assisted percutaneous interventions. The procedure of distal occlusion and aspiration prolongs the interventional time at the expense of faster early patency during unprotected interventions. Intuitively, however, one is tempted to speculate that a reduction or complete prevention of peripheral embolization is a mechanism by which distal flow is better preserved, and thereby may positively affect early and late outcome.^11,12 Moreover, it is possible that removal of potentially detrimental vasoactive substances from the culprit lesion or its vicinity, or their release during PTCA/stenting¹³ may be of advantage, but this aspect needs further study. In none of our patients did we encounter the no-reflow phenomenon when using the protection device and in only two patients did we observe minor peripheral embolizations to the distal portion of unprotected sidebranches. Branching in proximity of a culprit lesion might lead to embolization into major sidebranches once the catheter is placed in the main vessel. This limitation might be overcome by placing a second protection device into the sidebranch; this was safely performed in one patient of this series.

Duration of distal occlusion was longer in our study as compared to previously published work in saphenous vein grafts.^7,8 Direct stenting, which reduced the duration of the procedure, was not possible in all instances. In particular, long lesions and presence of intracoronary thrombi resulted in longer procedure times as did repeated aspirations. However, anginal symptoms during inflation of the device’s balloon were never intolerable and never caused hemodynamic impairment or rhythm disturbances requiring premature deflation.

In this primary series, we were able to demonstrate the feasibility and the safety of the PercuSurge GuardWire™. However, a large-scale trial with clinical endpoints is underway to investigate whether the strategy of aspiration with prevention of peripheral embolization improves outcome and which patients might be the best candidates for this procedure. Currently, there is still some debate as to directly compare percutaneous interventions with and without protection devices; in our opinion, however, and in light of the excellent results achieved so far and the clear body of evidence by the magnitude of the retrieved aspirates, we encourage interventionalists to apply this technique on a large scale.

Study limitations. We did not perform quantification or blinded analysis of flow characteristics after the intervention; this was only assessed semiquantitatively since it was the focus of this study to assess the safety and feasibility of the device when used in native coronary arteries.

In contrast to published work in saphenous vein grafts, the aspirates were semiquantitatively analyzed only; thus, no precise data are available on particle size of the debris. This will certainly be one of the important aspects of any future study with this protection device in native coronary arteries.

Conclusion
This preliminary experience with protection device-assisted interventions demonstrates the feasibility and safety of this system when used in native coronary arteries and distal occlusion by the device is well tolerated. The risk of peripheral embolization can be reduced and particles of sometimes remarkable size can be retrieved. Thus, to the extent that this material is involved in contributing to the mechanisms of distal embolization, no-reflow and infarction during percutaneous interventions, this device might contribute to reduce such complications.

Acknowledgments. We thank Mrs. Märki and Mrs. Sixt for their help and patience in preparing the photographs.

1. Kaplan BM, Larkin T, Safian RD, et al. Prospective study of extraction atherectomy in patients with acute myocardial infarction. Am J Cardiol 1996;78:383–388.
2. Litvack F, Eigler N, Margolis J, et al. Percutaneous excimer laser coronary angioplasty: Results in the first consecutive 3,000 patients. The ELCA Investigators. J Am Coll Cardiol 1994;23:323–329.
3. Nakagawa Y, Matsuo S, Kimura T, et al. Thrombectomy with AngioJet catheter in native coronary arteries for patients with acute or recent myocardial infarction. Am J Cardiol 1999;83:994–999.
4. Rosenschein U, Gaul G, Erbel R, et al. Percutaneous transluminal therapy of occluded saphenous vein grafts: Can the challenge be met with ultrasound thrombolysis? Circulation 1999;99:26–29.
5. Murakami T, Mizuno S, Takahashi Y, et al. Intracoronary aspiration thrombectomy for acute myocardial infarction. Am J Cardiol 1998;82:839–844.
6. Oesterle SN, Hayase M, Baim DS, et al. An embolization containment device. Cathet Cardiovasc Intervent 1999;47:243–250.
7. Carlino M, De Gregorio J, Di Mario C, et al. Prevention of distal embolization during saphenous vein graft lesion angioplasty. Experience with a new temporary occlusion and aspiration system. Circulation 1999;99:3221–3223.
8. Webb JG, Carere RG, Virmani R, et al. Retrieval and analysis of particulate debris after saphenous vein graft intervention. J Am Coll Cardiol 1999;34:468–475.
9. Chesebro JH, Knatterud G, Roberts R, et al. Thrombolysis in Myocardial Infarction (TIMI) Trial, Phase 1: A comparison between intravenous tissue plasminogen activator and intravenous streptokinase. Clinical findings through hospital discharge. Circulation 1987;76:142–154.
10. Rentrop KP, Cohen M, Blanke H, Philips RA. Changes in collateral filling immediately after controlled coronary artery occlusion by an angioplasty balloon in human species. J Am Coll Cardiol 1985;5:587–592.
11. Erbel R, Heusch G. Coronary microembolization — Its role in acute coronary syndromes and interventions. Herz 1999;7:558–575.
12. Topol EJ, Yadav JS. Recognition of the importance of embolization in atherosclerotic vascular disease. Circulation 2000;101:570–580.
13. Bates ER, McGillem MJ, Beals TF, et al. Effect of angioplasty-induced endothelial denudation compared with medial injury on regional coronary blood flow. Circulation 1987;76:710–716.