Approaches to Mechanical Coronary Thrombectomy

Donald S. Baim, MD
Donald S. Baim, MD

Against an atherosclerotic background of progressive accumulation of mural lipid, the episodic acute exacerbations of coronary artery disease (acute myocardial infarction [MI] and unstable angina) invariably involve the deposition of platelet-fibrin (white) thrombus, as well as classic (red) and mixed thrombus. The therapeutic approach to stabilizing such acute syndromes may be as simple as initiating effective anti-platelet therapy (aspirin, clopidogrel and platelet glycoprotein IIb/IIIa receptor blocker) or the addition of an anti-thrombin (heparin or low molecular weight heparin), before sending the patient for diagnostic angiography and stent implantation to treat the underlying atherosclerotic obstruction(s). But when the acuity of the syndrome warrants immediate intervention, and/or when a larger thrombus is present, the risk of simply proceeding with the mechanical intervention tends to increase the incidence of significant clinical problems (distal embolization, no reflow, abrupt vessel closure), and may lead to poor mural transfer from drug-eluting stents (DES).
However, it may be difficult to recognize small thrombi or to distinguish them from ulcerations in the underlying atherosclerotic substrate, although certain angiographic and clinical clues may be useful. Coronary thrombi are most likely in clinical settings where there has been a recent onset of symptoms, and angiographic findings of a mobile, rat-tail filling defect. The extent of the associated thrombus, however, may not become evident until a total occlusion is crossed with a guidewire, or even dilated with an under-sized balloon so that distal contrast flow is re-established to outline the extent of the filling defect. The final proof is that the filling defect is removed by spontaneous lysis (endogenous plasmin, working overtime in an anticoagulant milieu), the use of an exogenous plasmin activator (i.e., a thrombolytic drug), or the use of an effective mechanical thrombectomy catheter (Figure 1).

Suction Thrombectomy

In very acute thrombus, simple syringe suction applied through a tubular catheter with adequate lumen diameter may be sufficient to aspirate some thrombi. Several such catheters are available and have been used in the management of thrombotic lesions, although they are not approved for that specific use and only a few smaller registries have been performed, with mixed results, in acute MI. Examples include the Export catheter (Medtronic, Inc, Minneapolis, Minnesota), the Rescue and now the Rio catheter (Boston Scientific, Natick, Massachusetts), and the Pronto catheter (Vascular Solutions, Inc. Minneapolis, Minnesota). The Rinspirator (Kerberos Proximal Solutions, Mountainview, California) adds simultaneous saline infusion through exit points proximal to the distal aspiration lumen to create turbulence that may improve the efficiency of thrombus removal compared to the laminar flow provided the simple tubular aspiration catheters.

Cut and Aspirate Devices

The mechanical cutting and aspirating transluminal extraction catheter (TEC,™ Boston Scientific) device was developed to perform atherectomy, but emerged as an early device for the treatment of thrombus-containing native coronaries and vein grafts.1 Unfortunately, its use was associated with problems with distal embolization and vessel injury, and it is no longer being marketed. More recently, the X-Sizer® Thrombectomy device (eV3 Inc., Plymouth, Minnesota) device has used a rotating helical auger at its tip, combined with luminal suction, to remove clot. In the X-Sizer for Treatment of Thrombus and Atherosclerosis in Coronary Interventions Trial (X-TRACT)2, there was no net benefit except for a reduction in large MI (CK-MB > 8 times normal) in a subgroup with large baseline thrombi. In the X-sizer in AMI for negligible embolization and optimal ST resolution (X AMINE ST) Trial3, 201 patients with thrombolysis in myocardial infarction (TIMI) 0–1 flow within 12 hours of acute MI treated with the X-Sizer had enhanced ST resolution > 50% (68% vs. 53%; p = 0.037), less distal embolization (2% vs. 10%; p = 0.033), but no significant difference in TIMI flow, myocardial blush grade or 6-month major adverse cardiac events (MACE). The X-Sizer device is not approved in the United States for coronary thrombectomy.

Venturi-Bernoulli Suction

A high-speed water jet can create a surrounding area of low pressure by the Bernoulli-Venturi effect. This principle was used in the Hydrolyzer™ (Cordis Corp., Miami Lakes, Florida), which had only limited European use in coronary arteries and vein grafts.4,5 The dominant clinical application of this principle has been the AngioJet® Rheolytic™ Thrombectomy System (Possis Medical, Inc. Minneapolis, Minnesota), in which saline is injected at high pressure by an external pump, traveling through a small steel hypotube within the catheter, and spraying retrograde across a small gap near the tip of the catheter.6,7
According to the Venturi-Bernoulli principle, a low- pressure region is created surrounding this jet, which pulls surrounding fluid (blood, thrombus and saline) into the catheter. In the original AngioJet LF-140 device, the jet source was located at the tip of the catheter and sprayed proximally into the catheter opening, allowing the jets to macerate thrombus into small particles and propel them back through the catheter lumen and out of the body. The contemporary XMI and XVG Catheters (Possis Medical, Inc.) encase the jets within the outer catheter shaft, which has pairs of low- pressure (suction) orifices close to the jet, and higher-pressure (source) orifices located slightly more proximally to create a circular vortex around the tip of the catheter. This vortex design is more efficient in clot removal, allowing a smaller, 4-French (Fr) device in both over-the-wire and rapid-exchange formats that can be delivered through 6-Fr guiding catheters over 0.014-inch guidewires. The larger 5 Fr XVG Catheter is better suited to achieve complete thrombus removal in large (< 5.0 mm) coronary arteries and saphenous vein bypass grafts or large peripheral arteries. The Vein Graft AngioJet Study (VeGAS I) was a 90-patient, multicenter registry that showed the AngioJet reduced the angiographically measured thrombus burden within native coronary arteries or saphenous vein bypass grafts by an average of 86%.8 The VeGAS II randomized 349 patients with angiographically evident thrombus to either the AngioJet Rheolytic Thrombectomy System or intracoronary urokinase infusion (minimum of 8 hours [then-current standard of care for thrombus]), showing a significantly improved 30-day survival free of MACE (death, MI, emergent bypass surgery, target lesion revascularization or stroke).9 The results of VeGAS I and VeGAS II were the basis for the FDA’s approval of the device in March 1999 to remove coronary thrombus.
The technique for use originally advocated was to cross the target lesion completely with the AngioJet and to activate the device during pullback. This was intended to confirm that the original blunt-tipped device could actually cross the lesion, and to allow capture of any debris liberated during device operation in the distal suction aperture. Newer generation AngioJet devices have a more tapered tip better configured to cross such lesions, and can capture liberated debris even when positioned proximal to the thrombotic lesion by virtue of the recirculation vortex. Moreover, the passage of the device across the lesion clearly carries some risk of clot dislodgement, making proximal-to-distal aspiration the currently preferred technique for removing large native thrombi in the acute MI setting.

Newer Thrombectomy Devices

Ultrasonic vibration can induce cavitation, which can fragment thrombus into small particulates. Acolysis™ Therapeutic Ultrasound Technology (Angiosonics, Inc., Morrisville, North Carolina) delivers therapeutic coronary ultrasound (41.9 kHz) to the tip of a 5 Fr catheter.10 In the Adjuvant Tamoxifen Longer Against Shorter (ATLAS) Trial,11 181 saphenous vein graft (SVG) lesions were randomized to receive Acolysis or abciximab. Acolysis was inferior to thrombolylsis, with angiographic procedural success in 63% of Acolysis patients versus 82% of abciximab patients (p = 0.008), with a higher incidence of 30-day MACE (25% with Acolysis and 12% with abciximab, p = 0.036), reflecting more frequent non-Q-wave MI (19.6% vs. 7.9%, p = 0.03). An ultrasonic OmniWave guidewire (OmniSonics Medical Technologies, Inc., Wilmington, Massachusetts) is now in clinical trials to produce similar cavitation thrombus disruption, having been approved for that purpose in thrombotic dialysis shunts.12
Cavitation can also be created by laser interactions with blood, creating a rapidly expanding (and then collapsing) photo-acoustic steam bubble.13 A miniature laser cavitation technology was developed by Lawrence Livermore Nation Laboratories and licensed to EndoVasix, Inc. (Belmont, California), which performed intracranial human testing for stroke in 2002. That technology has now been acquired and is being developed for reintroduction into clinical testing by Selva Medical, Inc. (Belmont, California). The main advantage is that high-efficiency optical fibers allow this device to be as small as 3 Fr and flexible enough to be delivered into intracranial and coronary vessels.

Important Characteristics of Mechanical Thrombectomy Devices

Mechanical thrombectomy devices should have several important characteristics. First, the device must be deliverable to the target lesion. Virtually all of the devices discussed above are in this category. Second, the device should be capable of performing complete thrombectomy. This is more likely with all devices in the soft fresh thrombi in acute MI than in more chronic SVG thrombi, but even in acute MI we have seen several cases where simple suction has left remaining filling defects that are then removed by a more potent device such as AngioJet (Figure 2). Third, the device should be quick and easy to set up and relatively free of complications such as dissection or perforation, which are difficult to manage or reverse. Other transient complications (such as bradycardia due to adenosine release by hemolysis during AngioJet use) may make the use of prophylactic pacing necessary in treating lesions supplying the right coronary or dominant circumflex territory, as with other devices such as the Rotablator. Finally, there should be a clear clinical benefit to the performance of thrombectomy.
In part, removal of the thrombotic filling defect may allow for better estimation of the length of the underlying atherosclerotic lesion, more accurate stent placement and better approximation of a drug-eluting stent to the vessel wall. But the holy grail for thrombectomy in the acute MI setting is a reduction in complications such as infarct extension or no-reflow related to downstream embolization of resident thrombus during balloon dilation or stent placement. It is not clear how this should be measured (corrected TIMI frame count, TIMI myocardial blush grade, speed/completeness of resolution in ST-segment elevation or reduction in final size of the infarct [by creatine kinase curve, nuclear imaging or MR gadolinium enhancement]).14 The fact that the 480 patients in the AngioJet Rheolytic Thrombectomy In Patients Undergoing Primary Angioplasty for Acute Myocardial Infarction (AiMI) Trial failed to show those benefits suggests that AngioJet thrombectomy should not be performed routinely in patients undergoing primary angioplasty for acute MI. But it should be noted that although half of the randomized patients had an occluded infarct-related artery on initial angiography, only 20–22% of patients had moderate or large thrombi. It is certainly possible that the results could have been more favorable had better up-front screening for the presence of large thrombi been performed (e.g., by wiring or small-balloon dilation to establish antegrade flow), the trial restricted to patients with large thrombi, the more contemporary proximal to distal aspiration technique been used, etc.
The other point of concern in AiMI, however, was the fact that the 30-day composite MACE endpoint (death, new Q wave MI, stroke, TLR) was higher in the thrombectomy group (6.7 vs. 1.7%, p < 0.01). Detailed review of the 30-day deaths in the AngioJet arm appears elsewhere in this supplement, concluding no explicit relationship to device use, as do several substantial single-center experiences, which show more favorable outcomes and far lower incidence of adverse events (in particular, death). Whether the use of the AngioJet with current techniques will improve clinical outcomes in the subset of AMI patients with angiographic thrombus, however, will not be answered definitively until the completion of the European JetStent trial.


In conclusion, the presence of large coronary thrombi complicates the performance of catheter-based coronary intervention, but can be treated by a variety of mechanical thrombectomy devices. The choice of a particular device should be based on the size and age of the thrombus, demonstrated device efficacy and procedural ease. Despite compelling theoretical arguments, the benefits of routine mechanical thrombectomy in the setting of acute MI have not been demonstrated, and thrombectomy should currently be reserved for patients with large angiographic thrombus burden.





  1. Dooris M, Hoffmann M, Glazier S, et al. Comparative results of transluminal extraction atherectomy in saphenous vein graft lesions with and without thrombus. J Am Coll Cardiol 1995;25:1700–1705.
  2. Stone G, Cox D, Low R, et al. Safety and efficacy of a novel device for treatment of thrombotic and atherosclerotic lesions in native coronary arteries and saphenous vein grafts: Results from the multicenter X-Sizer for treatment of thrombus and atherosclerosis in coronary applications trial (X-TRACT) study. Catheter Cardiovasc Interv 2003;58:419–427.
  3. Lefevre T, Garcia E, Reimers B, et al. X-sizer for thrombectomy in acute myocardial infarction improves ST-segment resolution: Results of the X-sizer in AMI for negligible embolization and optimal ST resolution (X AMINE ST) trial. J Am Coll Cardiol 2005;46:246–252.
  4. van den Bos AA, van Ommen VG, Corbeij HM. A new thrombosuction catheter for coronary use: Initial results with clinical and angiographic follow-up in seven patients. Cathet Cardiovasc Diagn 1997;40:192–197.
  5. van Ommen VG, van den Bos AA, Pieper M, et al. Removal of thrombus from aortocoronary bypass grafts and coronary arteries using the 6 Fr Hydrolyser. Am J Cardiol 1997;79:1012–1016.
  6. Weisenaunt B, Baim D, Kuntz R, et al. Rheolytic thrombectomy with the Possis AngioJet — Technical considerations and initial clinical experience. J Invasive Cardiol 1999;11:421–426.
  7. Nakagawa Y, Matsuo S, Kiura 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.
  8. Popma J, Carrozza J, Ho K, et al. One-year outcomes for the Vein Graft AngioJet Study I (VeGAS I) Pilot Trial. Circulation 1997;96:216-I–217-I.
  9. Kuntz R, Baim D, Cutkip DJ, et al. A trial comparing rheolytic thrombectomy with intracoronary urokinase for coronary and vein graft thrombus (the Vein Graft AngioJet Study [VeGAS 2]). Am J Cardiol 2002;89:326–330.
  10. Rosenschein U, Roth A, Rassin T, et al. Analysis of coronary ultrasound thrombolysis endpoints in acute myocardial infarction (ACUTE trial). Results of the feasibility phase. Circulation 1997;95:1411–1416.
  11. Singh M, Rosenschein U, Ho K, et al. Treatment of Saphenous Vein Bypass Grafts With Ultrasound Thrombolysis: A Randomized Study (ATLAS). Circulation 2003;107:2331–2336.
  12. Smikahl J, Yeung D, Wang S, et al. Alteplase stability and bioactivity after low-power ultrasonic energy delivery with the OmniSonics resolution system. J Vasc Interv Radiol 2005;16:385–389.
  13. Dahm JB, Ebersole D, Das T, et al. Prevention of distal embolization and no-reflow in patients with acute myocardial infarction and total occlusion in the infarct-related vessel: A subgroup analysis of the cohort of acute revascularization in myocardial infarction with excimer laser-CARMEL multicenter study. Catheter Cardiovasc Interv. 2005;64:67–74.
  14. Kuntz RE, Rogers C, Baim DS. Percutaneous coronary intervention-induced emboli during primary PCI for STEMI: Too little, too much, or too late? Am Heart J 2005;150:4–6.