Impella CP Dislodgment, Swap, or Removal: Current Practices

Edo Kaluski, MD;  Safi U. Khan, MD;  Sudhakar Sattur, MD;  Dan Sporn, MD;  Maninder Singh, MD

Edo Kaluski, MD;  Safi U. Khan, MD;  Sudhakar Sattur, MD;  Dan Sporn, MD;  Maninder Singh, MD

Abstract: The Impella CP (ICP) catheter (Abiomed) offers hemodynamic superiority over the intra-aortic balloon pump. However, device-specific issues are occasionally encountered, especially when long-term mechanical circulatory support (MCS) is required. These include ICP dislodgment, ICP mechanical failure, and the need to remove the ICP while maintaining arterial access to either insert a new MCS device or to perform suture-based arteriotomy site closure. We offer a case in which ICP-based MCS involved ICP distal dislodgment, kinking of the ICP catheter, and removal of the device in a coagulopathic and thrombocytopenic patient. The literature regarding potential solutions to these problems is reviewed. Future ICP design is likely to reduce the frequency and complexity of such events and facilitate the management of these scenarios. This is the first report of a novel method to address ICP kinking dislodgment that did not respond to conventional repositioning maneuvers. 

J INVASIVE CARDIOL 2019;31(1):36-40.

Key words: mechanical circulatory support, thrombocytopenia

There is considerable increase in the use of the Impella CP catheter (ICP; Abiomed) for mechanical circulatory support (MCS) in cases of cardiogenic shock (CS) or complex percutaneous coronary intervention (PCI). We report a case of ICP dislodgment that required device repositioning and subsequent device removal in a patient with severe thrombocytopenia and coagulopathy. We discuss potential solutions to address the following management issues: (1) ICP catheter kinking dislodgment into the ascending aorta; and (2) the need to remove the ICP catheter while securing arterial access to the arteriotomy site. 

Case Report

A 63-year-old male with no previous cardiac history was found unresponsive and pulseless by his neighbor, who initiated chest compressions and contacted emergency medical services (EMS). The patient’s “down time” was uncertain. EMS found him in ventricular fibrillation (VF); after cardiopulmonary resuscitation (CPR), the patient had return of spontaneous circulation. On arrival to the emergency department, echocardiography revealed severe left ventricular dysfunction and ejection fraction of 10%, with clear wall-motion abnormalities most pronounced in the left anterior descending (LAD) artery territory. The patient was taken to the cardiac catheterization laboratory, where transradial coronary angiography revealed discrete single-vessel coronary artery disease (95% stenosis of the proximal LAD), which was treated with 2.75 x 12 mm Synergy drug-eluting stent (Boston Scientific), resulting in TIMI 3 flow. While the patient was initially stable, he deteriorated hemodynamically over the next few hours and was hypotensive on high doses of norepinephrine. He failed to oxygenate on 100% oxygen and suffered from countless events of ventricular fibrillation. The patient also developed recurrent symptomatic episodes of bradycardia and acute anuric renal failure. He was taken back to the cardiac catheterization laboratory for insertion of an ICP, a temporary pacing wire, and a pulmonary artery catheter. These tasks were accomplished uneventfully, and the patient returned to the coronary care unit for observation. After a few hours, it appeared that the ICP distal tip migrated into the ascending aorta. This was confirmed by ICP hemodynamics, echocardiography, and chest x-ray. The cardiology team attempted ICP repositioning at bedside using echocardiographic and fluoroscopic guidance; however, the tip of the ICP catheter was kinked and did not allow advancement of the catheter under fluoroscopic guidance into the left ventricle. The patient was brought back to the cardiac catheterization laboratory. 

Under full heparinization, the left radial artery was accessed with a 6 Fr radial sheath that was exchanged over a wire to a 70 cm x 6 Fr sheath. A 6 Fr Judkins right 4 guiding catheter was advanced over a 0.035˝ wire into the aortic arch. A 35 mm Amplatz GooseNeck snare (ev3) was advanced through the hemostatic Y-adapter and the guiding catheter to the aortic arch and captured the tip of the ICP device (the ICP pigtail) and pulled the ICP distal tip into the left subclavian artery to unfold the device (Figure 1). After unfolding the ICP tip, we could still not advance the ICP into the left ventricle. At that point, we elected to use the guiding catheter to lead the pigtail into the left ventricle – initially into the ascending aorta (Figure 2) and then into the left ventricle (Figure 3). Finally, the snare was released, the ICP maintained its position in the left ventricle (Figure 4), and full MCS was resumed. 

A day after repositioning the ICP and resuming MCS, the patient’s hemodynamics improved dramatically, and his left ventricular ejection fraction normalized. He remained in acute oliguric renal failure and developed profound anemia, thrombocytopenia, and coagulopathy. The patient was typed and crossed for replacement blood products and received a transfusion of packed cells, platelets, and fresh-frozen plasma. We elected to remove the ICP in the cardiac catheterization laboratory. After withdrawing the ICP catheter into the abdominal aorta, protamine was given to reverse anticoagulation. In view of the design of the ICP repositioning sheath, we used the previously described Trojan-Horse method,1 followed by a double Proglide (Abbott) closure of the arteriotomy site. In view of the thrombocytopenia and coagulopathy, we elected to use adjunctive contralateral wiring from the left femoral artery and imaging of the arteriotomy site after closure (Figure 5). 


This case report describes the following management issues occurring in a single patient requiring prolonged MCS with ICP: (1) distal dislodgment; (2) kinking of the distal tip of the ICP; and (3) need for ICP catheter removal while maintaining arterial access to allow suture-based arteriotomy site closure or to use the arterial access for alternative or similar MCS device. 

It appears that these ICP-related issues are not exceedingly rare, especially when prolonged MCS is required; however, neither the literature nor the industry can provide accurate estimates of the incidence. As in most cases, there was no apparent cause for catheter displacement; however, ICP dislodgment during CPR,2 cardioversion, defibrillation, patient movement, and even with no activity at all has been described. Similarly, device failure,3 hemolysis, thrombocytopenia, bleeding around the access site, and vascular complications during sustained mechanical circulatory support are not negligible, even with the lower-flow smaller sheath size of the ICP 2.5.4  

Preparing for prolonged MCS. The Abiomed package insert suggests that for prolonged MCS, the peel-away 14 Fr sheath (13 or 25 cm length) should be removed and a graded reposition sheath (9 Fr x 2 cm tapered tip with 15 Fr x 6 cm trunk) not equipped with side-port should be advanced. The current sheath design does not allow reinsertion of a wire into the arteriotomy site during device removal. Preclosure with 1-2 Proglides can be done; however, concern about suture sterility emerges and hence this strategy is not recommended by the manufacturer and does not allow convenient repeat access to the arteriotomy site. 

Abiomed redesigned a more refined graded repositioning sheath (with a 9 Fr x 2 cm tapered tip and a 15 Fr x 10 cm trunk) that is currently under limited release (Figure 6). This new sheath design harbors a wire side-port compatible with 0.035˝ wires. This design simplifies ICP removal while maintaining arterial access to allow reinsertion of a large-bore sheath (≥14 Fr) or late suture-based arteriotomy closure. 

There are at least two potential solutions to mitigate this problem when subjecting patients to prolonged MCS: 

(1)    Use of a 14 Fr non-ICP sheath, such as the 14 Fr x 13 cm Cook sheath (Cook Medical) instead of the ICP peel-off sheath or the sliding tapered-tip ICP sheath. Ideally, the shortest possible sheath should be used unless there is extreme vessel calcification or tortuosity. Using a conventional sheath will allow closure of the arteriotomy site post procedure while applying external manual pressure on the arteriotomy site. The advantage of the 14 Fr commercial sheath is that the ICP catheter can be removed or replaced entirely through the sheath in case of mechanical failure or device dislodgment. The disadvantage is that the larger external distal diameter and longer sheath is more likely to be occlusive to the downstream arterial circulation. 

(2)    Use of an arteriotomy site 0.018˝ parallel wire. After imaging the arteriotomy site, two wires are advanced via 5-6 Fr femoral sheath into the iliofemoral system toward the aorta. The ICP stiff, J-tip, 0.035˝ wire (or any other 0.035˝ stiff wire) and a 45 cm, 0.018˝ wire can be obtained from either a micropuncture kit or a radial sheath kit. The ICP peel-away sheath is advanced only over the 0.035˝ wire, while the 0.018˝ buddy-wire external tip is covered by the dilator of a 5 Fr micropuncture kit and is secured by a capped rotated stopcock (Figure 7) that is secured to the skin by a suture. In case the ICP needs to be removed or exchanged, it is simply removed with its ICP sheath while applying manual pressure, while the 0.018˝ buddy wire allows the advancement of a 6 Fr radial or femoral sheath under manual pressure exchange of the 0.018˝ wire to a 0.035˝ wire, which will allow insertion of a 14 Fr sheath or arteriotomy-site closure with 1-2 Proglides.

Removal of the ICP without the above advance preparations. The literature describes three major principal methodologies to remove an ICP catheter while maintaining arterial access (detailed in Table 1). Briefly, the first is the “Trojan-Horse” method. The second method is various iterations of cutting the ICP catheter at the 100 cm mark, removal of the repositional sheath, and using the severed ICP catheter to advance a longer sheath 9 Fr telescoped within a shorter 11 Fr5  or 12 Fr6  sheath, or a 10 Fr sheath telescoped within a 14 Fr sheath,7 or a distally severed ICP 14 Fr peel-away sheath.8 The third method is pulling the ICP back, exposing the device blood outlet area, and advancing a 0.018˝, 190 cm wire from the outlet orifice into the inlet orifice and up into the aorta to maintain access while removing the ICP catheter. A fourth potential solution is snaring the tip of the ICP catheter tip from the contralateral femoral, withdrawing the ICP catheter, and externalizing the snare and snare catheter as a rail to maintain femoral arteriotomy patency. 

Contralateral or ipsilateral wiring of the arteriotomy site prior to arteriotomy-site closure is optional to allow “dry seal” in the former (by using an occlusive balloon inflation) and to enable adequate imaging and treatment of arteriotomy-site complications. 

Repositioning of the dislodged ICP catheter. Unkinking or repositioning an ICP catheter is described here for the first time in the medical literature using a snare within a guiding catheter from the left radial approach. Previous reports of ICP kinking8 dislodgment were treated with ICP removal and insertion of a new Impella8 or another MCS.6 Since the cost of a second ICP is prohibitive and because removal of an ICP via the ICP tapered tip repositioning sheath is not without issues, the interventionist should make every effort to unfold or unkink and reposition the ICP device, assuming that MCR support is still needed and device functionality can be restored.


Dislodgment and need to remove an ICP while maintaining arterial access are events than can occur during prolonged MCS using the ICP. While the latter is being addressed by redesigned ICP repositioning sheaths, it is beneficial to acquire the knowledge and skills to mitigate or address these events. 


1.    Phillips CT, Tamez H, Tu TM, Yeh RW, Pinto DS. Novel method for exchange of Impella circulatory assist catheter: the “Trojan Horse” technique. J Invasive Cardiol. 2017;29:250-252 (Epub 2017 May 15).

2.    Aggarwal S, Bannon S. Displacement of Impella post chest compressions. HeartViews. 2014;15:127-128.

3.    Lemaire A, Anderson MB, Lee LY, et al. The Impella device for acute mechanical circulatory support in patients in cardiogenic shock. Ann Thorac Surg. 2014;97:133-138. Epub 2013 Oct 1.

4.    Lauten A, Engström AE, Jung C, et al. Percutaneous left-ventricular support with the Impella-2.5-assist device in acute cardiogenic shock: results of the ICP-EUROSHOCK-registry. Circ Heart Fail. 2013;6:23-30. Epub 2012 Dec 4.

5.    Yager N, Waxman S. Omran J, Reeves R. Techniques of Impella removal while preserving arterial removal with access site salvage. Catheter Cardiovasc Interv. 2018 Feb 26 (Epub ahead of print).

6.    Omran J, Reeves R. Techniques of Impella removal while preserving arterial access. Cardiovasc Revasc Med. 2018 Apr 6 (Epub ahead of print).

7.    Kim MS, Clegg S, Messenger JC. Removal of Impella 2.5 while maintaining vascular access: a solution to a vascular quandary. Catheter Cardiovasc Interv. 2014;83:223-225. Epub 2013 Sep 30. 

8.    Cook BS, Wilson C, Kaiser B, Baljepally R. A method for maintaining vascular access when Impella exchange is required. Catheter Cardiovasc Interv. 2017;90:945-947. Epub 2017 May 22.

From the Division of Cardiovascular Disease, Robert Packer Hospital and the Guthrie Health Services, Sayre, Pennsylvania and The Geisinger Commonwealth School of Medicine, Scranton, Pennsylvania. 

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.

The authors report that patient consent was provided for publication of the images used herein.

Manuscript submitted October 9, 2018 and accepted October 23, 2018.

Address for correspondence: Edo Kaluski, MD, FACC, FESC, FSCAI, Guthrie Robert Packer Hospital, 1 Guthrie Square, Sayre, PA 18840. Email: