Background. A series of thrombectomy and distal filter devices have been developed to limit distal embolization during percutaneous coronary interventions (PCI). Objective. To evaluate the feasibility of the combined use of thrombus-aspirating catheters and distal filter devices in patients at high risk of no-reflow. Methods. Thrombus aspiration (TA) and distal filter protection (DFP) were sequentially used in a series of patients undergoing urgent PCI within 48 hours of acute myocardial infarction (MI). Inclusion criteria were: (1) occlusion of the infarct-related artery; (2) at least 2 out of the 6 Yip’s classification features of high thrombus burden. Coronary angiograms were evaluated off-line to assess thrombus score, coronary flow and distal embolization in different phases of the procedure. Results. TA followed by DFP prior to balloon dilatation or stent implantation was successfully performed in 20 patients with acute MI due to occlusion of de novo lesions (80%) or in-stent thrombosis (20%) located in a native coronary artery (90%) or a saphenous vein graft (10%). TA was associated with a significant acute reduction of TS and improvement of coronary flow (TIMI grade from 0.7 ± 0.8 to 1.6 ± 1.1; p = 0.004 and CTFC from 83 ± 29 to 52 ± 36; p = 0.006). This facilitated the deployment of DFP, which did not induce significant flow modification (TIMI grade: 2.3 ± 0.9 pre-DFP placement vs. 2.2 ± 1.0 post-DFP placement; p = 0.20; CTFC: 32 ± 28 pre-DFP placement vs. 35 ± 28 post-DFP placement; p = 0.47). Post-PCI angiography revealed a 90% TIMI 3 flow rate and 47% MBG 3 rate with only 1 case of angiographically evident distal embolization. Conclusions. Sequential use of TA and DFP may be successfully used during PCI in patients at very high risk of distal embolization. However, the possible benefits of such an approach should be weighted with the increased complexity of the procedure. Further evaluations of the clinical efficacy of this approach are needed. J INVASIVE CARDIOL 2007;19:317–323

       Distal embolization of atherothrombotic material often occurs during percutaneous coronary intervention (PCI) in patients with acute myocardial infarction (AMI), and its angiographic evidence ifigures associated with worse long-term outcomes.¹ Angiographic features of a high thrombus burden also are predictive of no-reflow and higher mortality,² thus suggesting that interventional techniques able to reduce thrombus burden may improve myocardial reperfusion. A series of thrombectomy and distal protection devices with different mechanisms of action has been, or is going to be, tested in clinical practice to reduce distal embolization during PCI. Although routine use of these devices is not recommended, the risk of noreflow by standard PCI might be unacceptably high in the subgroup of patients with a high thrombus burden. We propose that the combination of thrombus aspiration (TA) followed by distal filter protection (DFP) can be useful in the management of highly thrombotic infarct-related arteries. Thus, in the present study, we assessed the feasibility and the angiographic results of TA followed by DFP during high-risk PCI.

Materials and Methods

       Study population. From September 2005 to August 2006, all consecutive patients with AMI and a single, occlusive, culprit lesion with “very high” thrombus burden undergoing urgent PCI within 48 hours from symptom initiation were selected to enter this pilot study.

       AMI comprised ST-elevation MI, which was defined as > 20 minutes’ duration of chest pain with > 0.1 mV ST-elevation in at least 2 contiguous leads. Non-ST-elevation MI was defined as typical chest pain with documentation of transient > 0.1 mV ST-segment depression or T-wave modifications in at least 2 contiguous leads with a ≥ 2-fold increase in serum creatine kinase-MB.

       The presence of a “very high” thrombus burden at the site of the culprit vessel was defined as the presence of ≥ 2 of the following Yip’s criteria for high thrombus burden²:

  1. Large infarct-related artery (visually estimated reference vessel diameter ≥ 4 mm)

  2. Angiographic thrombus with the greatest linear dimension > 3 times the reference vessel diameter;

  3. “Cutoff pattern” (lesion morphology with an abrupt cutoff without taper before the occlusion);

  4. Accumulated thrombus (> 5 mm of linear dimension) proximal to the occlusion;

  5. Floating thrombus proximal to the occlusion;

  6. Persistent dye stasis distal to the obstruction.

       PCI with combination of thrombus aspiration and distal filter protection. PCI was performed via the right radial or the femoral approach. After placement of a 6 Fr guidingcatheter (shape chosen by the operator), a 0.014 inch guidewire was advanced through the culprit lesion. Next, TA was performed with the Diver CE catheter (Invatec, Brescia, Italy) by slowly advancing it in aspiration through the culprit lesion, as previously described.³ When the 30 ml syringe was full (and the aspiration force was consequently abolished), the device was retracted (in aspiration) in the guiding catheter. This procedure was performed 1 to 6 times, depending upon particulate yield and immediate angiographic results. Generally, TA was performed with the aim of restoring anterograde flow and creating a tunnel within the thrombotic lesion that was able to allow lesion crossing with the closed distal filter device. If, after TA, the operator was not confident that sufficient lumen to place the distal filter device had been gained, predilatation with an undersized balloon (ratio ≤ 0.5:1 with the reference vessel diameter) was performed. A distal filter device was then deployed. The DFP devices used were the FilterWire EZ^™ (Boston Scientific Corp., Natick, Massachusetts) or the Spider^™ (ev3, Inc., Plymouth, Minnesota), according to the operator’s discretion and estimated reference vessel diameter (since the FilterWire device for small vessels was not available, only the Spider filter was used for vessels < 3.5 mm). The characteristics as well as the mode of employ of the FilterWire EZ⁴ and Spider filter have been previously described.⁵ The two filter devices have different designs, however, they have similar pore size in the capture basket (110 microns) and comparable clinical efficacy when used to treat saphenous vein grafts.⁵ As previously reported,⁶ FilterWire EZ deployment was facilitated in the present study by keeping, as a buddy wire, the 0.014 inch guidewire used for initial lesion crossing and TA. When the FilterWire EZ was located in the desired distal location, the buddy wire was removed, and the wire of the filter remained the operative guidewire. Thereafter, the procedure was continued at the discretion of the operator with the aim of achieving the best possible angiographic result. In de novo lesions, a bare-metal stent was implanted with the aim of fully covering the culprit lesion, and postdilatation with a different balloon was needed, in some cases, to optimize stent apposition to the wall. Additional stent implantation was sometimes required to cover minor dissections and plaque shifts, or to treat other significant lesions in the same vessel. When the culprit lesion was an in-stent thrombosis, multiple inflations with balloons were used to dilate the stent, and additional stent implantation was used only to cover dissections out of the stent. At the end of the procedure, the distal filter device was closed and captured with the appropriate retrieval catheter and then removed.

       All patients were treated with heparin (initial weightadjusted intravenous bolus followed by additional boluses administered, with the aim of obtaining an activated clotting time of 250–300 seconds) and with double antiplatelet therapy with aspirin and clopidogrel (loading dose of 600 mg followed by 75 mg/day) for at least 4 weeks. Abciximab (0.25 mg/kg bolus plus infusion of 0.125 μg/kg/minute for 12 hours) was administered unless contraindicated. Beta-blockers, ACEinhibitors and statin drugs were administered as appropriate in the absence of any specific contraindication. After the procedure, patients underwent repeated sampling (every 8 hours for 2 days, then daily) for cardiac enzyme assessment (troponin T, creatine kinase-MB, creatine kinase).

       Angiographic analyses. Coronary angiograms at the following stages of the PCI procedure were reviewed off-line by two expert interventional cardiologists (ER and MDV):

  1. Before intervention;

  2. After guidewire crossing of the target lesion;

  3. After TA;

  4. Before distal filter placement;

  5. After distal filter placement;

  6. Before distal filter removal (after stenting or after last balloon dilatation in patients not treated with stenting);

  7. At the end of PCI.

       Anterograde coronary flow was graded using the standard thrombolysis in myocardial infarction (TIMI) criteria.⁷ Corrected TIMI frame count (CTFC) was measured according to Gibson.⁸ Thrombus score (TS) was graded as previouslydescribed by the TIMI study group⁹ (0 = no thrombus; 1 = haziness; 2 = definite thrombus < 1/2 vessel diameter; 3 = definite thrombus 1/2 to 2 vessel diameters; 4 = definite thrombus > 2 vessel diameters; 5 = occluded vessel). Features of high thrombus burden were categorized according to Yip’s classification.² Myocardial blush grade (MBG) at the end of the procedure was evaluated according to van't Hof et al.¹⁰ Distal embolization was defined as occlusion with an abrupt “cut-off” appearance at angiography of a branch of the infarctrelated artery distal to the culprit lesion site.¹

       Statistical analysis. Continuous variables (presented as mean ± standard deviation) were compared by paired Student’s t-test or Wilcoxon and Mann-Whitney U-tests, as appropriate. Chi-square tests (Fisher’s corrected, when appropriate) were used to compare discrete variables (reported as raw numbers [%]). Analyses were carried out using SPSS for Windows, version 11.0 (SPSS, Chicago, Illinois). Statistical significance was defined by two-tailed p < 0.05. Filter no-reflow was defined as a CTFC acute > 10 units improvement.¹¹

Results

       Study population. Out of the 287 patients with acute ST-elevation or non-ST-elevation MI undergoing urgent (< 48 hours) PCI during the study period, 21 (7.3%) consecutive patients were selected for treatment with sequential TA and DFP.

Table 1

Clinical and preprocedural angiographic characteristics of the study.

  

       One patient, a 37-year-old female with a recent diagnosis of uterine carcinoma, presenting with anterolateral ST-elevation MI (STEMI) complicated by cardiogenic shock, was found to have distal left main thrombosis with total occlusion of the left anterior descending (LAD) artery and subocclusive stenosis of the ostial circumflex artery (CX) (with angiographic evidence of spontaneous distal embolization on the distal CX). In this patient, it was not possible to establish any flow in the LAD with TA as well as with balloon dilatations, and TA + DFP were successfully used in the left main CX axis with the aim of obtaining hemodynamic stabilization.

Figure 1B

(B) (black arrows underscore the high thrombus burden involving all the vessel), after thrombus aspiration with a Diver CE

Figure 1A

Example of sequential thrombus aspiration and distal filter protection in a patient with occlusive, late (8 months) in-stent thrombosis of a long drug-eluting stent (Taxus® 3.5 x 28 mm) implanted in the mid-right coronary artery. The vessel is represented before percutaneous coronary intervention (PCI) (A), after guidewire placement

 

Unfortunately, after achieving initial stabilization, the patient died after 48 hours of recurrent irreversible cardiogenic shock. In this particular case, the main culprit lesion was the left main LAD, which was not reopened regardless of the technique adopted (TA and balloon dilatation), while the technique was successfully applied on the side branch (i.e., the CX), thus rendering it impossible to consider the patient in the angiographic analyses.

Accordingly, 20 patients comprised the study population. Their detailed clinical and angiographic characteristics are reported in Table 1. Briefly, most of the patients presented with ST-elevation MI and long symptom-to-cath lab time, abciximab was used in 85% of the cases. Baseline angiographic features are summarized in Table 1.   

       PCI with sequential thrombus aspiration and distal filter protection. As previously described, the planned strategy was to first perform TA with the aim of reducing thrombus burden and creating a tunnel sufficient to receive the crossing profile of the selected DFP device, then, to protect the distal circulation with the DFP device and perform a protected angioplasty with or without stenting.

Figure 2B

(B) Material captured within the basket of the FilterWire EZ™ in the same patient.

Figure 2A

(A) Material obtained with thrombus aspiration in the patient with occlusive, late (8 months) in-stent thrombosis of a long drug-eluting stent (Taxus® 3.5 x 28 mm) implanted in the mid-right coronary artery.

 

Figure 1 shows the angiography in the different steps of a procedure and Figure 2 shows the aspirated and distally captured material obtained in the same patient). The planned sequence of TA followed by DFP was modified by introducing unplanned balloon dilatations in the following circumstances: (1) in 6 cases (30%), the wiring phase was difficult and prompted the operator to perform predilatation with a 1.25–1.5 mm diameter balloon to confirm the correct intraluminal positioning of the guidewire before advancing the TA catheter; (2) in 8 cases (40%), the angiographic result after TA was not considered satisfactory and the operator decided to perform predilatation with an undersized balloon (balloon:vessel ratio < 0.5:1) to allow easy positioning of the DFP device.

       The TA phase of the procedure was not associated with angiographically evident distal embolization and induced a significant acute reduction of TS (from 4.3 ± 0.6 to 3.5 ± 1.1; p = 0.0034) and improvement of TIMI grade and CTFC (TIMI grade: from 0.7 ± 0.8 to 1.6 ± 1.1; p = 0.004 and CTFC: from 83 ± 29 to 52 ± 36 post-TA; p = 0.006). Macroscopically visible aspirated material was retrieved by TA in 16 /20 (80%) of the patients.

Figure 2B

(B) Material captured within the basket of the FilterWire EZ™ in the same patient.

Figure 2A

(A) Material obtained with thrombus aspiration in the patient with occlusive, late (8 months) in-stent thrombosis of a long drug-eluting stent (Taxus® 3.5 x 28 mm) implanted in the mid-right coronary artery.

       The placement of the primary selected DFP device was successful in all but 1 case (a patient with late stent thrombosis inwhom the FilterWire EZ was not able to cross the stent; further balloon predilatation was performed, and a Spider 4 mm device was successfully placed distal to the stent). The comparison of angiograms obtained pre- versus post-DFP placement showed angiographically evident distal embolization in only 1 case (5%), and an overall absence of significant flow modification (TIMI grade: 2.3 ± 0.9 pre-DFP placement vs. 2.2 ± 1.0 post-DFP placement; p = 0.20; CTFC: 32 ± 28 pre- DFP placement vs. 35 ± 28 post-DFP placement; p = 0.47).

       After establishment of DFP, the PCI procedure was performed according to the operator’s discretion (Table 2). When the treatment of the culprit lesion was considered completed, the DFP device was removed by using the dedicated retrieval catheter. This phase is usually a critical one, and guiding catheter support was pivotal (the advancement of the retrieval catheter through vessel angulations and long, stented segments is often difficult).

Figure 1F

Figure 1E

(E) (white arrow indicates the opened filter), and after PCI

 

In 3 cases, this phase was particularly problematic, requiring deep guiding catheter intubation, and in 1 case, it was not possible to properly advance the retrieval catheter using the standard technique. In the latter patient, the retrieval catheter of a Spider 4 mm filter was not able to cross a long, stent-covered (46 mm of total stent length), angulated segment of a large right coronary artery (RCA). This situation, which has been previously described to be a cause of urgent cardiac surgery,¹² was solved by advancing an extra-support Choice PT “buddy wire” beside the Spider’s wire and by mounting and advancing the retrieval catheter over the two wires (filter’s wire + buddy wire), achieving retrieval catheter crossing of the stent. Finally, the filter was closed by advancing the 6 Fr guiding catheter through the stent via deep intubation.

       Comparison of angiograms obtained before and after DFP removal showed an overall improvement of coronary blood flow (TIMI grade flow from 2.6 ± 0.5 to 2.8 ± 0.5; p = 0.055; CTFC from 21 ± 10 to 15 ± 7; p = 0.0005), probably expressing an overall mechanical obstacle offered by the material captured by the filter cage.

Table 2

Technical characteristics of the procedures and angiographic results obtained.

 

Accordingly, in 7 patients (35%) a filter no-reflow phenomenon was observed. Macroscopically visible captured material was present in 13 (65%) of the filters.

       The final results obtained in the study population are reported in Table 2, and the overall behavior of coronary flow (as evaluated by CTFC) observed in the study population is graphically summarized in Figure 3. After the procedure and throughout hospitalization, no patient died or had reinfarction, and discharge was performed after 7 ± 3 days. Only 1 patient underwent target vessel revascularization; he was treated in the setting of inferior STEMI of the RCA, and was scheduled for elective PCI of the LAD after 48 hours. Even if the clinical course was uneventful and no further cardiac enzyme rise was observed on serial assessments, control angiography of the RCA showed distal occlusion at the site of filter placement during the previous PCI. Successful PCI with stenting was then performed.

Figure 3

Mean value (with standard deviation) of corrected TIMI frame count in the different phases of the percutaneous coronary intervention.

Discussion

       Distal embolization often occurs during PCI in patients with thrombotic lesions, and its occurrence is associated with adverse outcomes.¹ Accordingly, a series of devices with theoretical anti-embolic properties have been developed and are actually being tested in clinical practice. Some randomized trials using thrombectomy and distal protection devices have provided conflicting results,¹³ so that the routine use of adjunctive devices is currently not recommended. However, angiographic features of high thrombus burden predict the occurrence of distal embolization^2,14 and have been clearly associated with suboptimal reperfusion in patients with AMI.^2,15 Not surprisingly, a prospective, randomized study¹⁶ showed that patients with more angiographically evident thrombosis constitute the subgroup that has a greater amount of material captured with a filter device¹⁴ and a greater reperfusion benefit with TA. Thus, it would appear that the use of adjunctive devices should be limited to patients with a higher risk of distal embolization. Even greater care in patient selection should be exercised when combining the use of different devices in different phases of the procedure. Accordingly, we decided to test the feasibility of combining TA and DFP in patients with higher angiographically evident coronary thrombosis. The study population quantitatively represented a minority of the patients with AMI treated in the study period (about 7%), and was selected on the basis of the simultaneous presence of at least two of the Yip’s angiographic criteria for “highburden thrombus formation.” The observed results are promising, as indicated by the high angiographic reperfusion rate, in spite of the adverse risk profile pattern. Indeed, the 90% rate of TIMI 3 flow compares favorably with the TIMI 3 rate ranging between 27% and 55% in the presence of only one of the Yip’s criteria in his series of 794 patients undergoing standard primary PCI.²

       Rationale for sequential use of thrombus aspiration and distal filter protection. The concept of combining thrombectomy with distal protection is not new. One of the available devices with documented clinical benefit in the treatment of degenerated saphenous vein grafts, i.e., the PercuSurge GuardWire (Medtronic AVE, Minneapolis, Minnesota), combines thrombus aspiration and distal protection properties. However, it has been shown that the use of this device does not improve the results of PCI in patients with AMI caused by obstruction of native coronary arteries.¹⁷ The presence of an occlusive, distal balloon protection system may be responsible for such negative results. Indeed, it increases ischemic time and causes deviation of coronary flow (and possibly embolic material) through side branches located between the lesion and distal protection device.

       To date, no available device combines TA with distal filter (nonocclusive) protection; we therefore decided to test the combination of two different devices. Even if the combined use of two devices increases costs, it offers the possibility of managing difficult cases with devices that have previously been usedindividually in less complicated situations, thus eliminating the need for complex device-related learning curves.

       If the concept of combining two different devices is accepted, the sequence of usage comes into discussion. Distal filters are loaded onto guidewires so that the most obvious sequence is to try to deploy distal filter protection as early as possible in order to perform the most protected PCI possible. Accordingly, this is the principal action of the PercuSurge GuardWire device. However, we hypothesized that distal filter placement may be associated with relevant distal embolization in occlusive thrombotic lesions due to the characteristics of the available devices (suboptimal flexibility and a high crossing profile, which is 2.9 Fr for the Spider and 3.2 Fr for the FilterWire EZ). We therefore tested a sequence of TA as the first step, aimed at reducing the thrombus burden and creating a channel which, in turn, might reduce the risk of distal embolization during filter placement. Since the aspiration lumen of simple TA catheters is about 3 Fr, they offer the opportunity to create a tunnel of sufficient diameter to accept the filter’s deployment catheter crossing profile. Moreover, their use in the first step allows restricted “provisional” usage of a distal filter only in cases of a residual, huge thrombus burden following TA.

       The present report shows that the described strategy is feasible and is associated with a progressive increase of anterograde flow during PCI.

       Emerging limitations of the combined use of thrombus aspiration and distal filter protection. Both the placement and the removal of distal filter devices in native coronary arteries appear to be technically challenging. In particular, the removal of distal filters may be associated with problems in crossing long, stented coronary segments or vessel tortuosities with the retrieval catheters. Surgical removal of uncaptured filters has been recently described by Limbruno as a complication directly related to filter usage.¹² Furthermore, in one of our cases, the operator was forced to advance the retrieval catheter on both the filter wire and a buddy wire to avoid surgery.

       Another possible limitation of the use of filters in native vessels may be their safety for endothelium of the distal infarct-related artery. Indeed, injury by the distal occlusion balloon of PercuSurge GuardWire has been shown to cause the development of stenosis,18 and we observed a case of silent vessel occlusion at the site of distal filter placement.  

Conclusions

       In conclusion, the sequential use of TA and DFP during PCI of high thrombus burden lesions is feasible and is associated with excellent angiographic results. Although the present study has several limitations including a small number of patients enrolled in a single center and its observational nature, it strongly supports the need for controlled, randomized trials to test the potential benefit of TA followed by DFP in selected high-risk patients.

       Acknowledgments. We would like to thank all the nurses and technicians at our catheterization laboratory for their invaluable help in the introduction of new devices to our daily practice. We also thank the young doctors working in the cath lab during their training in cardiology for their collaboration in the systematic collection of the clinical and technical data.