Use of the Impella 2.5 in High-Risk Percutaneous Coronary Intervention

In Europe, the Impella 2.5 has been in wide use for several years.^8,9  In the United States, several small trials^5,6,9–12  have been completed or are underway to evaluate the Impella 2.5. Abiomed has sponsored several of these trials, including PROTECT I, PROTECT II, RECOVER I, RECOVER II, and ISAR-SHOCK.

The PROTECT I study was done to assess the safety and efficacy of device placement in patients undergoing high-risk PCI.¹⁰  Twenty patients were enrolled in that prospective, multicenter study. All had poor ventricular function (ejection fraction =35%) and had PCI on an unprotected left main coronary artery or the last remaining patent coronary artery or graft. The device was successfully placed in all patients, and the duration of support ranged from 0.4 to 2.5 hours. None of the patients had hemodynamic instability develop during the procedure. Two patients had transient hemolysis that was not clinically significant. Following this trial, the Impella 2.5 device received its 510(k) approval from the Food and Drug Administration.

The ISAR-SHOCK trial was done to evaluate the safety and efficacy of the Impella 2.5 versus the IAPB in patients with cardiogenic shock due to acute myocardial infarction.⁵  Patients were randomized to support from an IABP (n=13) or an Impella (n=12). One patient died before implantation of a device. The trial’s primary end point of hemodynamic improvement was defined as improved cardiac index at 30 minutes after implantation. Improvements in cardiac index were significantly greater in the patients with the Impella than in patients with the IABP (P=.02). The patients’ diastolic pressure increased significantly more with Impella support than with IABP support (P=.002). Mean arterial pressure increased in patients with an Impella more than it increased in patients with an IABP, but the difference was not significant (P=.09). The use of inotropic agents and vasopressors was similar in both groups of patients. Serum level of lactate was lower in patients treated with the Impella. Hemolysis, as measured by the plasma level of free hemoglobin, was higher in patients treated with the Impella. More blood products were administered to patients treated with the Impella.

Results from additional clinical trials are available (Table 2), but continuing clinical trials are essential to evaluate the effectiveness of the Impella 2.5 device in various subgroups.

The Impella 2.5 catheter (Figure 2) contains a nonpulsatile microaxial continuous flow blood pump (Figure 3) that pulls blood from the left ventricle to the ascending aorta, creating increased forward flow and increased cardiac output. An axial pump is one that is made up of impellar blades, or rotors, that spin around a central shaft; the spinning of these blades is what moves blood through the device.¹³  The distal tip of pigtail curve is 6F in size (Figure 4). The Impella 2.5 catheter shaft size is 9F and is 12F at its largest point, which is the microaxial blood pump.

The Impella 2.5 catheter has 2 lumens. A tubing system called the Quick Set-Up has been developed for use in the catheterization laboratory. It is a single tubing system that bifurcates and connects to each port of the catheter. This arrangement allows rapid initial setup of the console so that support can be initiated quickly. When the Quick Set-Up is used, the 10% to 20% dextrose solution used to purge the motor is not heparinized. One lumen carries fluid to the impellar blades and continuously purges the motor to prevent the formation of thrombus. The proximal port of this lumen is yellow. The second lumen ends near the motor above the level of the aortic valve and is used to monitor aortic pressure. The console uses this pressure measurement to calculate the catheter’s position. The proximal port of this lumen is red.

If the Impella remains in place for continued support after the PCI, the tubing system is changed to 2 separate systems. For continued use, the purge fluid is heparinized (50 units heparin per milliliter of fluid) and infused via an integrated infusion pump, the Braun Vista infusion pump (Figure 5). The typical infusion rate for the purge fluid is 7 to 20 mL/h.

The second lumen of the Impella 2.5 is flushed with heparinized saline (500 mL normal saline with 1000 units heparin) by using regular (not infusion pump) intravenous tubing. The heparinized saline is placed in a pressure bag that is inflated to 300 mm Hg. Other components of the Impella 2.5 catheter include the cable that connects the catheter to the console and a repositioning sheath for bedside repositioning of the Impella 2.5.

The components required to run the device are assembled on a rolling cart (Figure 6) and include the power source, the Braun Vista infusion pump, and the Impella console (Figure 7). The Impella console powers the microaxial blood pump and monitors the functioning of the device, including the purge pressure and several other parameters. The console has 10 function keys, but not all of the function keys are used for the Impella 2.5. Some were designed to support future versions of Impella catheters. Function keys that are used with the Impella 2.5 include the P-PERF, MENU, SIGNAL, SCALE, and the ON keys. The console can run on a fully charged battery for up to 1 hour.

The Impella 2.5 catheter is placed percutaneously through the common femoral artery and advanced retrograde to the left ventricle over a guidewire. Fluoroscopic guidance in the catheterization laboratory or operating room is required. After the device is properly positioned, it is activated and blood is rapidly withdrawn by the microaxial blood pump from the inlet valve in the left ventricle and moved to the aorta via the outlet area, which sits above the aortic valve in the aorta (Figure 8).

If the patient tolerates the PCI procedure and hemodynamic instability does not develop, the Impella 2.5 may be removed at the end of the case while the patient is still in the catheterization laboratory. The Impella device can be withdrawn, leaving the arterial sheath in place. The sheath can then be removed later when the patient’s activated clotting time or partial thromboplastin time has returned to near normal levels. For patients who become hemodynamically unstable or who have complications during the PCI (eg, no reflow, hypotension, or lethal arrhythmias), the device can remain in place for continued partial circulatory support, and the patient is transported to the critical care setting.

The Impella 2.5 cannot be used on all acutely ill patients who require hemodynamic support.¹⁴  Because the device is designed to sit across the aortic valve in the left ventricle, it should not be used in patients who have prosthetic aortic valves, so as to prevent damage to the valve. It should not be used in patients with moderate to severe aortic insufficiency; it may worsen the degree of insufficiency because the aortic valve cannot close completely with the device in place. The Impella should not be used in patients with a heavily calcified aortic valve. The device may cause calcium to break off of the aortic valve and embolize, causing stroke. It cannot be used in patients with documented left ventricular thrombus. The impellar blades of the motor could break up the thrombi, causing arterial embolization. The device also cannot be used in patients with severe peripheral arterial disease because of the large sheath size required for its placement, nor can it be used in patients who cannot tolerate anticoagulation with heparin.

The most commonly reported complications of Impella 2.5 placement and support include limb ischemia, vascular injury, and bleeding requiring blood transfusion.^6,9  Hemolysis has been reported.^5,10  Other potential complications include aortic valve damage, displacement of the distal tip of the device into the aorta, infection, and sepsis. Device failure, although not often reported, can occur. Potential complications of use of the Impella 2.5 are detailed in Table 3.

The nursing care of patients remaining on Impella 2.5 support requires astute assessment and expert nursing care (Table 4). Frequent hemodynamic evaluation is needed. The use of continuous cardiac output monitoring may be useful for patients with cardiogenic shock. Inotropic agents, such as dobutamine and milrinone, and vasopressors, such as dopamine and norepinephrine, may still be needed after the Impella 2.5 is placed to maintain a cardiac index of at least 2 and systolic blood pressure at 90 mm Hg or higher. Some patients on Impella support may be intubated and receiving mechanical ventilation. Implementation of the ventilator bundle is required for these patients, including elevation of the head of the bed to decrease the risk of ventilator-associated pneumonia, as well as deep venous thrombosis and peptic ulcer prophylaxis. Regular repositioning of patients and assessment of their skin are necessary to prevent skin breakdown.

The patient with an Impella 2.5 is at risk of limb ischemia because of the large size of the device. The arterial access site must be assessed regularly for bleeding and development of a hematoma. Distal pulses of the affected leg should be assessed at least hourly. The use of anticoagulation is required, and bleeding may develop in some patients. The partial thromboplastin time should be monitored every 4 to 8 hours and maintained at about 45 to 55 seconds, unless a higher partial thromboplastin time is required for some other reason.

Hemolysis can occur in patients who are on the Impella 2.5. Hemolysis can be mechanically induced when red blood cells are damaged as they pass through the microaxial pump. When hemolysis occurs, hemoglobin level and hematocrit decrease, haptoglobin level decreases, and plasma levels of free hemoglobin increase. If the hemolysis is severe, hemoglobinuria will be seen; if the hemolysis is sustained, acute kidney injury will occur.

Patients on Impella 2.5 support who may require interrogation of a permanent pacemaker or implantable cardioverter defibrillator present an interesting situation. In order for the interrogator to connect with the permanent pacemaker or implantable cardioverter defibrillator, the Impella console must be turned off for a few seconds while the signal is established. As soon as the signal has been established, Impella support is immediately restarted.

The recommended maximum performance level for continuous use is P8. At P8, the flow rate is 1.9 to 2.6 L/min and the motor is turning at 50000 revolutions per minute. When activated, the console is silent. No sound other than alarms is audible during Impella support, unlike the sound heard with an IABP. Ten different performance levels ranging from P0 to P9 are available (Table 5). The performance level determines the flow rate and the number of revolutions per minute. As the performance level increases, the flow rate and number of revolutions per minute increase. At maximum performance (P9), the pump rotates at 50000 revolutions per minute and delivers a flow rate of 2.1 to 2.6 L/min. P9 can be activated only for 5-minute intervals when the Impella 2.5 is in use.

Several parameters require regular monitoring for the duration of therapy. Regular documentation is essential, and review of trends is useful when troubleshooting. Parameters to assess regularly include placement signal, placement monitoring (pump position), dual signal, purge pressure, motor current, and speed (Table 6).

Several console alarms may require nursing intervention (Table 7). The most common include suction, low purge pressure, and high purge pressure alarms. Information about the alarm can be seen on the console screen. An audible alarm also will sound. Suction alarms can occur if the performance level is too high for the patient; for example, in a patient who is hypovolemic or if the device is emptying the ventricle. A low purge pressure alarm indicates that the purge pressure to the Impella motor has decreased below 300 mm Hg. A low purge pressure can allow blood to enter the motor and damage the motor, rendering the device inoperable. Low purge pressures require immediate intervention by a critical care nurse. A high purge pressure alarm indicates that the purge pressure is greater than 700 mm Hg and generally indicates a kink in the tubing or catheter.

If the device advances too far and both the inflow and outflow areas are fully in the left ventricle, the pump position wrong alarm will occur. On the placement signal screen, the waveform will show a ventricular, rather than an aortic, pressure waveform. The motor current signal will be flattened. The placement monitoring diagram will show the aortic valve symbols distal to the pump (Figure 9). The performance level should be decreased to P2 and the physician should be notified to reposition the device by pulling it back slightly to obtain an aortic waveform. Repositioning is best performed under fluoroscopic guidance, whenever possible.

If the Impella 2.5 is completely in the aorta, or if the inlet and outlet areas are in the ventricle and the pressure port is in the aorta, the pump position alarm will sound. The placement signal will show a normal appearing aortic waveform with systolic and diastolic pressures similar to those shown by the patient’s arterial catheter. The motor current will be flattened. The placement monitoring diagram will continuously scroll through 3 different images because it is not able to differentiate the position of the catheter. The performance level should be reduced to P2 and the physician should be notified of the need to reposition the Impella.

If the patient’s ventricular function is very poor, a positioning alarm can occur. The alarm will read “pump position unknown due to low pulsatility.” The Impella console is not able to determine the pump position because the patient’s systolic and diastolic pressures are not very different. The placement signal will show depressed cardiac function, evidenced by a damped appearing waveform. The motor current will be flattened. The aortic valve symbols will not be seen on the placement monitoring diagram.

Weaning from the partial circulatory support provided by the Impella 2.5 can be approached in different ways. When the device has been in only for support during high-risk PCI and the patient’s condition is stable after the procedure is completed, rapid weaning can be done. For rapid weaning, the performance level is decreased by 1 or 2 levels every 5 to 15 minutes until the pump is set at P2. The patient’s hemodynamic status is assessed after every decrease in performance level. P2 is the lowest performance level that can be used while the distal end of the Impella 2.5 is in the left ventricle. At the end of rapid weaning, the Impella device can then be removed as described in the next section.

Patients who have had the device in longer or who were in unstable condition during the procedure may benefit from a slower weaning process. Slow weaning is accomplished by decreasing the performance level by 1 or 2 levels every 2 or 3 hours. If the patient tolerates this slow weaning by remaining hemodynamically stable while at P2, the device is then removed.

In our facility, we often use transthoracic echocardiography to aid in the assessment of our patients. With the physician at the bedside, transthoracic echocardiography is performed and the pump’s performance level is decreased. The physician can evaluate left ventricular size and function and can quickly assess how well the patient tolerates decreases in the performance level.

When the physician is ready to discontinue the Impella 2.5 catheter, the groin dressing should be removed, the site cleansed with chloroprep, and the sutures clipped. Immediately before removal of the device, decrease the performance level to P0. The catheter is then pulled back from the left ventricle into the ascending aorta and then completely withdrawn. The arterial sheath can be left in place to be removed separately, or the catheter and the sheath can be removed as a unit, as is common when removing an intra-aortic balloon catheter placed through a sheath.

The Impella 2.5 instructions for use recommend a 40-minute manual hold to achieve hemostasis at the arterial access site. We follow our hospital’s protocol for achieving hemostasis, using either manual or mechanical compression. Some facilities may use what is referred to as a preclose method while the patient is still in the catheterization laboratory; that technique uses 2 8F Perclose suture-mediated closure devices (Abbott Laboratories, Abbott Park, Illinois). The Impella 2.5 can also be removed in the surgical setting with direct visualization of the common femoral artery. A cutdown is performed, and the device is removed by the surgeon. Arterial repair is done at that time, and the incision is covered with a sterile dressing.

To date, we have placed an Impella 2.5 in about 40 patients. Our facility has a high-volume interventional cardiology program as well as a busy adult and pediatric cardiovascular surgery service. We have extensive experience with the IABP and with surgically placed VADs, including the Abiomed AB5000, Thoratec CentriMag, XVE, and HeartMate II. Our use of the Impella 2.5 was our first experience with percutaneously placed partial circulatory support devices. We have also placed the Impella 5.0 in 3 patients. We have gained much knowledge since implementing our Impella program in April 2008. Based on our experiences, we share the following recommendations.

1. Assign a point person who can oversee and coordinate the program. In our facility, this role is filled by the interventional cardiology clinical nurse specialist. All patients with this device are observed by this clinical nurse specialist during their hospital stay. Additionally, patients are observed by the VAD coordinator.

2. Limit the physicians who can place and manage the device and have them proctor one another. In our facility, physicians who can place this device are those who have interventional cardiology and/or cardiac surgery privileges. All cases are also followed by the cardiac surgeon who oversees our VAD program and a critical care intensivist. We developed preprinted orders for physicians to use for patients remaining on support to ensure consistency in care (Table 8).

3. Train a core group of nurses and cardiovascular technologists from the catheterization laboratory for initial setup of the console, catheter preparation, and assisting with placement of the device. Console and infusion pump setup requires 2 to 3 minutes to complete. Our training for staff from the catheterization laboratory consisted of a 3-hour session with didactic and hands-on training. Ongoing practice is essential to develop and maintain competency. The Impella coordinator is present for implants to assist the catheterization laboratory staff as needed. The VAD coordinators also provide valuable assistance. We found it helpful for the first few Impella placements to be planned, elective placements for high-risk PCI. This arrangement allowed ample time for setting up the console and infusion pump before the Impella was placed.

4. Develop a nursing protocol that outlines key strategies in patient management and a specific flow sheet (Figure 10) for documentation.

5. Train a core group of critical care nurses to care for the patient, monitor the device, change tubings, and troubleshoot alarms. We provide 1-to-1 staffing for our patients with an Impella 2.5 until they are hemodynamically stable. Our training for our cardiac intensive care unit (CICU) nurses consisted of a series of three 1-hour sessions, encompassing a review of pertinent physiology and hemodynamics, Impella console management and troubleshooting, and tubing changes with hands-on practice and documentation practice. Staff received 1-to-1 mentoring from the Impella coordinator. Review of patient care management strategies, console troubleshooting, and fluid/tubing setup has also been added to our annual VAD skills fair (Table 9).

6. Work with pharmacy and nursing staff to develop a standardized time for fluid and tubing changes. We found performing such changes at a consistent time to be very helpful. This arrangement ensured that new purge fluid was ready when needed and that the nurses had more staffing resources to assist with the tubing change. The pharmacy prepares the heparinized 20% glucose purge solution and sends it to the unit before our standard change time of 3 pm daily. We have 2 nurses at the bedside for the tubing change. The Impella coordinator is present to assist with tubing changes as needed by the nursing staff.

7. Include the cardiovascular surgery staff and perfusion staff in the initial training. Patients with an Impella may go to the cardiovascular operating room for bypass surgery or valve repair/replacement. Coordination with the perfusionist is essential when patients are being placed on cardiopulmonary bypass. As the patient goes on cardiopulmonary bypass, the performance level of the Impella must be decreased. When the patient is coming off of cardiopulmonary bypass, the performance level of the Impella pump, if the pump remains in place, must be increased to provide adequate flow to the patient. In some facilities, the perfusionist manages both the cardiopulmonary bypass and the Impella. In our facility, we have a trained CICU nurse managing the Impella while the patient is in the cardiovascular operating room.

8. Partner with industry. Abiomed provides strong clinical support as well as excellent print and Web-based educational materials. A console simulator is available that can be used to practice troubleshooting. Abiomed also provides assistance with troubleshooting via 24-hour-a-day, 7-day-a-week telephone support.

The Impella 2.5 device is an alternative to an IABP that may be beneficial in some patients who need hemodynamic support. It is useful in patients undergoing high-risk PCIs. The Impella 2.5 should not be used in patients with aortic valve disease or known left ventricular thrombus. Potential complications include bleeding, vascular injury, hemolysis, infection, sepsis, and device malfunction or failure. Nursing care is directed toward support of critically ill patients, including frequent hemodynamic assessment, titration of vasoactive medications as indicated, Impella console troubleshooting, and monitoring for potential complications. A multidisciplinary approach to establishing a program for the Impella is useful in ensuring competency and good outcomes for patients. The following 3 case studies demonstrate the successful use of Impella 2.5 support during high-risk PCIs at our facility.

A 71-year-old man was referred to our facility for saphenous vein bypass graft (SVG), a high-risk PCI. Diagnostic cardiac catheterization and coronary angiography performed at the referring facility showed severe triple vessel coronary disease and high-grade lesions in 2 SVGs. He had a history of coronary artery bypass grafting surgery several years prior as well as heart failure. His estimated ejection fraction was markedly decreased at 20% (normal, 55%–70%) on a recent echocardiogram. He was experiencing progression of his anginal symptoms, and his functional capacity was poor. He declined repeat bypass surgery. After careful evaluation and optimization of pharmacological therapy for heart failure, the decision was made to proceed with high-risk PCI of the SVG to the left anterior descending (LAD) artery with an Impella 2.5 used to provide partial circulatory support during the PCI.

The patient was brought to the catheterization laboratory and prepared for the procedure. On arrival, his blood pressure was 151/88 mm Hg with a heart rate of 66/min. Sheaths were placed in the right femoral vein, the right femoral artery, and the left femoral artery. An SvO₂ pulmonary artery catheter showed that the patient’s baseline pulmonary artery pressures were markedly elevated at 69/40–47 mm Hg (normal: 20–30/8–12 mm Hg; mean, 25 mm Hg). Heparin 7000 units was administered intravenously for anticoagulation. When the activated clotting time was higher than 250 seconds, the Impella 2.5 was advanced into position via the left common femoral artery and placed across the aortic valve into the left ventricle. The catheter device was connected to the Impella console and support was initiated. The performance level was increased from P0 to a maximum of P8 in a matter of minutes. At P8, the flow was 2.5 L/min and the patient’s cardiac index was 2.58 (normal, 2.5–4).

The PCI procedure of the LAD SVG was then started. The SVG was aspirated before a distal protection device was placed. One bare metal stent (not a drug eluting stent) was placed in the SVG to the LAD. During this time, transient no-reflow developed. No reflow is a phenomenon in PCI that can cause severe hemodynamic compromise because myocardial perfusion through a given segment of the coronary circulation is inadequate, yet no angiographic evidence of mechanical obstruction of the vessel is apparent.¹⁵  The patient’s pulmonary artery pressures increased to 93/40–58 mm Hg and the cardiac index decreased to 2.2. Despite the no-reflow, the arterial pressure, heart rate, and cardiac rhythm remained stable, and the patient had no complaints of shortness of breath or chest discomfort. Adequate blood flow was reestablished after the intracoronary administration of 500 μg nitroglycerin. Because the patient was tolerating the procedure well after this, the physician decided to stent the second diseased SVG. With the Impella continuing at maximum flow of 2.5 L, 2 bare metal stents were deployed in the SVG to the posterior lateral branch. Pulmonary artery pressures remained elevated after this stent was placed, and the patient was given furosemide and nitroglycerin to decrease preload and increase diuresis. Just before the patient’s return to the CICU, the pulmonary artery pressure decreased to 70/41–48 mm Hg and the cardiac index was 2.69.

The patient was weaned from the Impella device, the device was removed in the catheterization laboratory, and the venous and arterial sheaths were secured in place. The total duration of Impella support was slightly less than 2 hours. The patient was transferred to the CICU for continued monitoring. Four hours later, after the activated clotting time was less than 150 seconds, the venous and arterial sheaths were removed and hemostasis was achieved after 40 minutes of manual compression.

The patient’s remaining hospital course was uneventful and he was discharged 2 days later taking appropriate medications for after PCI and to treat heart failure. Six weeks later, the patient was pain free with no complaints of dyspnea. Transthoracic echocardiography showed increased contractility, and the ejection fraction was 40%. At follow-up 12 months later, the patient remained symptom-free with a normal ejection fraction of 55%.

A 74-year-old man was transferred to our facility after sustaining a non–ST-segment elevation myocardial infarction. He had a long history of diabetes and had undergone coronary artery bypass surgery 20 years prior. Diagnostic cardiac catheterization showed severe triple vessel disease as well as severe diffuse distal disease of his left internal mammary artery (LIMA) graft, his only remaining patent graft. He had severe mitral regurgitation. His ventricular function was poor with a calculated ejection fraction of 20%. His left ventricular end-diastolic pressure was markedly elevated at 40 mm Hg (normal, 5–12 mm Hg), and his cardiac index was reduced at 1.98. He was evaluated by our cardiothoracic surgery team and was deemed an extremely poor candidate for surgery. After a family conference that included the patient, the decision was made to proceed with high-risk PCI of the LIMA graft, using the Impella 2.5 for hemodynamic support during the procedure.

He returned to the catheterization laboratory the next day for coronary intervention. On arrival, his blood pressure was 105/66 mm Hg with a heart rate of 64/min. Sheaths were placed in the right femoral vein, the right femoral artery, and the left femoral artery. Heparin 9000 units was administered intravenously for anticoagulation, and the Impella 2.5 was advanced into position in the left femoral artery. Support was established at P8. The coronary guidewire was advanced through the LIMA graft to the distal LAD. Three sequential stents were placed in the area of disease. The patient’s vital signs and cardiac rhythm remained stable throughout the procedure. The patient was rapidly weaned from the Impella, which was removed in the catheterization laboratory. The 13F arterial sheath was sutured in placed for removal 4 hours later after the patient was returned to the CICU. He was discharged home 2 days later.

A 63-year-old woman was admitted to our facility with an acute anteroseptal myocardial infarction. She was treated with fibrinolytic therapy but continued to have chest pain. She had a history of hypertension complicated by end-stage renal disease and was receiving dialysis treatments 3 times per week. She was taken to the catheterization laboratory, where she was found to have a significant obstruction of the left main coronary artery complicated by plaque rupture. She was evaluated by our cardiac surgery team, who thought that she was a very high-risk candidate for surgery based on the recent fibrinolytic therapy, her renal failure, and the remote history of stroke. She was transferred to the CICU and supportive medical therapy was initiated.

She returned to the catheterization laboratory 3 days later for high-risk PCI of the left main coronary artery. On arrival, her blood pressure was 127/72 mm Hg, and her heart rate was 54/min. Sheaths were placed in the right femoral vein, the right femoral artery, and the left femoral artery. Bivalirudin was administered for anticoagulation, and the Impella 2.5 was advanced into position in the left femoral artery. Support was established at P8.

The left main lesion was crossed, and one stent was deployed at the lesion. During stent deployment, the patient had severe chest pain and showed ST-segment elevation in the anterior leads. Despite these changes, her blood pressure and cardiac rhythm remained stable. Intravascular ultrasound was performed to confirm optimal stent placement. At the end of the procedure, she was symptom free, and she was rapidly weaned from the Impella and the device was removed. The 13F arterial sheath was sutured in placed for removal 2 hours later, after the patient had returned to the CICU. She had no further chest pain during her hospitalization and was discharged home in stable condition 3 days later.