Cardiogenic shock carries high in-hospital mortality despite invasive interventions.1 Short-term mechanical circulatory support systems have been explored for the management of acute cardiogenic shock. Percutaneous left ventricular assist devices (LVAD) are catheter-guided devices placed across the aortic valve into the left ventricle to pump blood from the left ventricle into the ascending aorta. This allows the device to immediately reduce strain on the left ventricle while continuing to increase overall systemic cardiac output.2,3 In the United States, there are several approved devices that can be chosen based on the individual’s specific cardiac needs, including flow rate, left-sided, or right-sided support.
There have been efforts to use percutaneous LVAD in acute decompensated heart failure and high-risk percutaneous coronary intervention (PCI).3,4 Clinical trials have compared percutaneous LVAD to intraaortic balloon pump in acute cardiogenic shock and high-risk PCI. A meta-analysis of randomized controlled trials of percutaneous left ventricular support devices showed superior hemodynamic parameters in comparison to intraarterial balloon pump (IABP) but without differences in 30 day mortality.4
Ischemic stroke and intracranial hemorrhage are common causes for morbidity and mortality in patients with long-term implantable LVAD,5–9 with many of these strokes occurring during the perioperative period.7 Incidence of acute ischemic and intracranial hemorrhages during the percutaneous, short-term cardiac support remains largely unknown. We aimed to study acute neurologic complications during percutaneous LVAD support.
Study Design and Population
We reviewed prospectively collected data of consecutive persons implanted with Impella devices (Abiomed Inc., Danvers, MA) at a single, tertiary center between October 2010 and July 2018. We included all adult patients (age ≥18) implanted with Impella devices 2.5, 5.0, CP, or RP. This study was approved by the local institutional review board. Strict compliance to the International Society for Heart and Lung Transplantation’s ethics statement is upheld at our institution.
All patients were extracted from a local, prospective database (Electronic Data Interface for Transplantation [EDIT]) which included preimplant demographics, medical history, social history, type of device, and clinical status variables, as well as postimplant clinical status variables and adverse event variables.6 Patients were routinely placed on therapeutic anticoagulation with unfractionated heparin infusion with a target activated partial thromboplastin times (aPTT) goal of 53–78 seconds. They were also continued on antiplatelet agents, including various dosages of aspirin 81–325 mg daily.
Acute neurologic events (ANE) in this study were classified as ischemic strokes or intracranial hemorrhages, including intracerebral hemorrhage, subdural hematomas, and subarachnoid hemorrhages. Additional data for patients with ANE were retrospectively reviewed and collected from the hospital electronic medical record including antithrombotic regimen, international normalized ratios (INRs), and aPTTs at the time of acute hemorrhagic event or ischemic stroke, stroke severity by National Institutes of Health Stroke Scale (NIHSS), time of known baseline before stroke, and time from implantation to neurologic event.8,9 All patients were followed from the time of implant until death. Information on withdrawal of care or mortality was collected.
All patients with ANE had computed tomography (CT) imaging of brain. Computed tomography images were evaluated by an independent neuroradiologist at the time of stroke. Both images and reports for all patients were reviewed by one investigator (C.E.H.). Radiographic variables such as infarct size, infarct location, and the location of occluded vessel were collected. We assessed ischemic infarct and intraparenchymal hemorrhage volume by using the ABC/2 method.10
The results are presented as median ± range with six individual cases discussed. Statistical comparisons on demographic and clinical characteristics were performed using Student’s t test Fisher’s exact test or Mann–Whitney U test. A p value less than 0.05 was considered significant.
Seventy-nine persons underwent Impella implantation between October 2010 and July 2018 with a median support of 8 days (range 1–33 days). Six (7.5%) persons with ANE included three ischemic strokes, two intracerebral hemorrhages, and one subdural hematoma. The median age was 58 years (range 42–78) with five (84%) males (Table 1). Of the six patients with ANE, the Impella 5.0 device was implanted in five persons (Table 2). Acute neurologic events occurred in a median of 5 days from Impella implant (range 1–8).
Table 1. -
Characteristics of Patients with Acute Ischemic Strokes and Intracranial Hemorrhages During Impella Implantation
|Demographics and Medical History
(N = 79)
(N = 73)
(N = 6)
|Age, median (range)
||63 (20 – 83)
||63 (20 – 83)
||58 (42 - 78)
|Male, N (%)
|Bridge to heart transplantation, N (%)
|Type of device
| Impella 5.0
| Impella CP
| Impella RP
|Duration of support in days, median (range)
|NYHA class IV preimplant, N (%)
|INTERMACS implant status, N (%)*
| Status not available
|Diabetes mellitus, N (%)
|Hypertension, N (%)
|Arrhythmia, N (%)
|Prior myocardial infarction, N (%)
|Prior ischemic stroke, N (%)
|Prior coronary artery bypass grafting, N (%)
|Prior percutaneous coronary intervention, N (%)
|Implantable cardiac defibrillator, N (%)
|Smoking, N (%)
|Alcohol, N (%)
*One Impella patient did not have an INTERMAC status assigned during their hospitalization.
INTERMACS, Interagency Registry for Mechanically Assisted Circulatory Support; NYHA, New York Heart Association.
Table 2. -
Clinical Information and Outcomes for the 6 Patients with Acute Neurologic Events During Percutaneous Left Ventricular Assist Device Implantation
||Implant to ANE (Days)
||Infarct or Hemorrhage Volume
||AT at the Time of ANE
||Thrombo cytopenic at the Time of ANE
||Mortality at Hospital Discharge
||Other Risk Factors at the Time of ANE
||Posterior fossa SAH, corpus callosum splenium ICH
||Aspirin 81 mg, heparin drip
||Yes (91 ×103/µL)
||Deceased, withdrawal of care
||Right temporoparietal ICH
||Aspirin 81 mg, heparin drip
||No (229 ×103/µL)
||Deceased, withdrawal of care
||Left middle cerebral artery distribution infarct
||Aspirin 162 mg, heparin drip
||Yes (82 ×103/µL)
||Deceased, withdrawal of care
||Posterior falx and left tentorial subdural hematoma
||Yes (55 ×103/µL)
||Living, heart transplantation achieved during admission
||Postoperative day 2
Orthotopic heart transplant
||Acute right occipital and left frontal infarct
||R Occipital: 4 ml
L Frontal: 1 cc
|Aspirin 300 mg
||Yes (113 ×103/µL)
||Decreased, PEA arrest
||Impella 5.0 1/9
Impella RP 1/30
|Subacute right frontal and left corona radiata infarcts
||R Frontal: 2 ml
L Corona radiata: 1 cc
||Yes (54 ×103/µL)
||Living, heart transplantation achieved during admission
ANE, acute neurologic event; aPTT, activated partial thromboplastin; AT, antithrombotics; CT, computed tomography; ICH, intracerebral hemorrhage; INR, international normalized ratio; PEA, pulseless electrial activity; SAH, subarachnoid hemorrhage.
Case No. 1: Posterior Fossa Subarachnoid Hemorrhage and Corpus Callosum Hemorrhage
A 67 year old male with idiopathic dilated cardiomyopathy (New York Heart Association Functional Classification [NYHA] IV) was hospitalized with acute exacerbation of chronic systolic heart failure complicated by nonsustained ventricular tachycardia. Impella 5.0 was placed for temporary hemodynamic support while awaiting destination therapy with LVAD. He was initiated on aspirin 81 mg and unfractionated heparin for maintenance. His hospital course was complicated by thrombocytopenia with platelet count ranging 70–90 ×103/µL.
On postimplant day 8, the patient was noted to be unresponsive with an initial NIHSS score of 24. Computed tomography imaging of brain depicted a right paracentral corpus callosum splenium hemorrhage with an extensive posterior fossa subarachnoid hemorrhage extending from the bilateral tentorial leaflets to the foramen magnum surrounding the upper cervical spinal cord (Figure 1). Computed tomography angiogram of head and neck was negative for vascular malformation or aneurysm. The family decided to withdraw active support and pursue comfort care.
Case No. 2: Right Temporoparietal Intracerebral Hemorrhage with Uncal Herniation
A 52 year old male with familial nonischemic cardiomyopathy presented in acute cardiogenic shock and admitted for pulmonary artery catheter-guided management, inotropic support, and IABP. Intraarterial balloon pump was replaced with Impella 5.0 on hospital day 7 as a bridge to transplant. On postimplant day 4, the patient had acute decline in alertness with left-sided weakness with an initial NIHSS score of 28. Computed tomography imaging of head depicted right temporoparietal intraparenchymal hemorrhage of approximately 80 ml with intraventricular extension (Figure 1). At the time of hemorrhage, platelet count was 266 ×103/µL and aPTT 73.2 sec. Patient received protamine reversal, mannitol, and 23.4% hypertonic saline without change in neurologic status. Repeat CT imaging of head 6 hours later depicted a new left occipital intraparenchymal hemorrhage (10 ml). His surrogate medical decision-makers decided to withdraw life-sustaining therapy.
Case No. 3: Left Middle Cerebral Artery Territory Infarction
A 48 year old male with a history of PCI with drug-eluting stent to the proximal left circumflex presented with myocardial infarction complicated by medically refractory atrial flutter. Left heart catheterization depicted severe three-vessel coronary artery disease requiring coronary artery bypass grafting. Intraoperative course was complicated by severe biventricular dysfunction requiring emergent Impella 5.0 placement. His postoperative course was complicated by vasoplegia, thrombocytopenia, and persistent lactic acidosis requiring inotropic support.
On postimplant day 2, he was found to be unresponsive during his first evaluation of sedation. Computed tomography imaging of head depicted an acute left middle cerebral artery infarct (~100 ml) (Figure 1C). The patient’s unfractionated heparin drip was stopped after the time of stroke diagnosis due to concern for hemorrhagic conversion. A weaning trial of the Impella device was unsuccessful. On resumption of the heparin infusion, his thrombocytopenia worsened from 82 to 40 ×103/µL. Postimplant day 3, he suffered nonconvulsive seizures requiring antiepileptic medications. Decision was made to withdraw life-sustaining therapy leading to discontinuation of Impella device and hospice care.
Case No. 4: Left Tentorial Subdural Hematoma
A 61 year old female with nonobstructive hypertrophic cardiomyopathy was admitted in acute cardiogenic shock and was found to have an acute reduction in left ventricular ejection fraction from 40% to 27%. Impella CP was placed via right axillary approach on admission day 39 and underwent heart transplantation on day 42 after 3 days of support. On removal of Impella at the time of heart transplantation, a right subclavian thrombus was found requiring intraoperative thrombectomy. Intraoperative course was also complicated by severe right ventricular dysfunction, vasoplegia, and lactic acidosis.
After failing to awaken off sedation on posttransplant day 1, CT imaging of brain was performed, which depicted a small, incidental 2 mm posterior falx and left tentorial subdural hematoma. Heparin drip had been discontinued before heart transplantation; however, coagulation tests at the time of intracranial hemorrhage diagnosis were elevated with an INR 1.8 and aPTT 62.4. Subcutaneous heparin was restarted on postbleed day 5 and aspirin on day 10. The patient was further medically stabilized and discharged to acute rehabilitation.
Case No. 5: Acute Right Occipital and Left Frontal Infarcts
A 73 year old male presented with a myocardial infarction and was found to have occlusion of the left circumflex and right coronary arteries with a 90% stenosis of the left anterior descending artery. He had an unsuccessful PCI to left anterior descending artery and was placed on extracorporeal membrane oxygenation (ECMO). He required intraaortic balloon pump and was weaned off ECMO with Impella 5.0 insertion via right axillary artery on admission day 3. Three days postimplant, CT imaging of brain obtained for increased lethargy showed two ischemic infarcts in the right occipital and left frontal lobes. Impella was removed after 8 days of support after his hemodynamics tolerated the assist device wean. His condition stabilized but developed pulseless electrical activity cardiac arrest on day 26 of hospitalization.
Case No. 6: Bilateral Frontal Lobe and Left Corona Radiata Infarcts
A 42 year old male was admitted for acute decompensated heart failure due to nonischemic cardiomyopathy. His hospital course was complicated by cardiogenic shock, acute liver, and renal failure. He underwent Impella 5.0 placement via right axillary approach and was initiated on unfractionated heparin infusion. Twenty-two days later, he required RP Impella placement for bi-Impella support with heparin discontinued during the implantation process. His RP Impella was removed 6 days after implantation for concern of pump thrombosis with a clot visualized via transthoracic echocardiogram on the tip of the Impella catheter. After RP Impella explant, he experienced an acute mental status change. CT imaging of brain showed infarcts in the right frontal lobe and left corona radiata. The etiology of the infarcts was deemed likely cardioembolic from the known Impella thrombus. The patient went on to successfully receive a heart transplant.
Impella is utilized to provide hemodynamic support during acute cardiogenic shock or high-risk percutaneous coronary artery intervention. The patients requiring Impella devices are inevitably high risk for stroke complications based on the acuity of their presentation. We found that 7.5% of persons with Impella support developed ischemic stroke and intracranial hemorrhage complications. This is in contrast to prior studies that found less than 2.0% incidence of hemorrhagic and ischemic strokes during the implantation period.11,12 Potential factors that may contribute include duration of support, inadequate anticoagulation, and thrombocytopenia.
For the cardiogenic shock period, Impella 2.5, CP, 5.0, and LD catheters are to be utilized as temporary ventricular support devices for short-term use, which is recommended for <4 days for Impella 2.5 and CP, and <6 days for Impella 5.0 and LD. In this case series, we found that the median time from implant to ANE was 5 days (Table 2). Several Impella studies have noted that the risk of complications, such as systemic hemorrhages, increase with longer usage beyond 4 days.13,14 Longer durations of support may also place patients at risk for ischemic strokes and intracranial hemorrhages.
Impella devices require a continuous purge solution that contains heparin to prevent pump thrombosis and device failure. Placing a thrombogenic pump device in a patient with the possibility for anticoagulation interruption inevitably further increases the risk of thromboembolism. While pump thrombosis is a known complication for LVAD devices, there are limited data with Impella.8 In our case series, patient no. 6 was found to have pump thrombosis likely leading to embolic strokes. In addition, all ischemic strokes within this case series had embolic patterns consistent with thromboembolism seen with other cardiac devices.7,8
While therapeutic heparin is necessary to keep the device functioning, this exposure also places patients at risk for hemorrhagic events. A three-center, retrospective study of acute decompensated heart failure patients that received Impella 5.0 as a bridge to decision found that access site bleeding requiring transfusion was the most common in-hospital complication.14 Our case series showed three intracranial hemorrhages occurred in the setting of recent heparin exposure with two patients (no. 1 and no. 2) experiencing fatal intracranial hemorrhages.
Thrombocytopenia is a known complication for Impella devices.3 Compared with Impella patients who survived to next therapy, a retrospective study found that patients who died on Impella support were found to have lower mean platelet counts.14 A randomized comparison of intraaortic balloon support and percutaneous LVAD also noted higher rates of severe bleeding complications due to disseminated intravascular coagulation, occurring among 62% of Impella patients compared to 15% of IABP patients.14 In our cohort, five patients with either ischemic or intracranial hemorrhagic were found to be thrombocytopenic at the time of ANE with a median platelet count of 87 k/UL (range 54–229).
Our study has limitations, including small sample size, single-center, and retrospective study design. It is difficult to determine the exact etiology for the neurologic events in the setting of acute cardiogenic shock as there are confounding factors. Other causes of strokes, such as atrial fibrillation, ventricular thrombus, atherosclerosis, and surgeries, may also have contributed other than the major factors we described. Other invasive procedures, such as the placement of other mechanical circulatory support devices, such as the presence of ECMO in case no. 5 may also have contributed to stroke risk. Further investigations with larger randomized control trials and prospective studies are warranted to further delineate predictors and preventative methods for these complications.
Ischemic stroke and intracranial hemorrhage appear to be common during percutaneous LVAD support. Further study is needed to characterize the risk factors and predictors of stroke in percutaneous LVAD.
1. Thiele H, Zeymer U, Neumann FJ, et al.; IABP-SHOCK II Trial Investigators: Intraaortic balloon support for myocardial infarction with cardiogenic shock. N Engl J Med 2012.367: 1287–1296.
2. Burzotta F, Trani C, Doshi SN, et al. Impella ventricular support in clinical practice: Collaborative viewpoint from a European expert user group. Int J Cardiol 2015.201: 684–691.
3. O’Neill WW, Kleiman NS, Moses J, et al. A prospective, randomized clinical trial of hemodynamic support with Impella 2.5 versus intra-aortic balloon pump in patients undergoing high-risk percutaneous coronary intervention: The PROTECT II study. Circulation 2012.126: 1717–1727.
4. Cheng JM, den Uil CA, Hoeks SE, et al. Percutaneous left ventricular assist devices vs. intra-aortic balloon pump counterpulsation for treatment of cardiogenic shock: A meta-analysis of controlled trials. Eur Heart J 2009.30: 2102–2108.
5. Kirklin JK, Naftel DC, Pagani FD, et al. Seventh INTERMACS annual report: 15,000 patients and counting. J Heart Lung Transplant 2015.34: 1495–1504.
6. Frontera JA, Starling R, Cho SM, et al. Risk factors, mortality, and timing of ischemic and hemorrhagic stroke with left ventricular assist devices. J Heart Lung Transplant 2017.36: 673–683.
7. Cho SM, Moazami N, Frontera JA. Stroke and intracranial hemorrhage
in HeartMate II and HeartWare left ventricular assist devices: A systematic review. Neurocrit Care 2017.27: 17–25.
8. Cho SM, Hassett C, Rice C, et al. What causes LVAD-associated ischemic stroke? Surgery, pump thrombosis, antithrombotics, and infections. ASAIO J 2019.65: 775–780.
9. Acharya D, Loyaga-Rendon R, Morgan CJ, et al. INTERMACS analysis of stroke during support with continuous-flow left ventricular assist devices: Risk factors and outcomes. JACC Heart Fail 2017.5: 703–711.
10. Kothari RU, Brott T, Broderick JP, et al. The ABCs of measuring intracerebral hemorrhage volumes. Stroke 1996.27: 1304–1305.
11. O’Neill WW, Schreiber T, Wohns DH, et al. The current use of Impella 2.5 in acute myocardial infarction complicated by cardiogenic shock: Results from the USpella Registry. J Interv Cardiol 2014.27: 1–11.
12. Ranc S, Sibellas F, Green L. Acute intraventricular thrombosis of an impella LP 5.0 device in an ST-elevated myocardial infarction complicated by cardiogenic shock. J Invasive Cardiol 2013.25: E1–E3.
13. Hall SA, Uriel N, Carey SA, et al. Use of a percutaneous temporary circulatory support device as a bridge to decision during acute decompensation of advanced heart failure. J Heart Lung Transplant 2018.37: 100–106.
14. Thiele H, Sick P, Boudriot E, et al. Randomized comparison of intra-aortic balloon support with a percutaneous left ventricular assist device in patients with revascularized acute myocardial infarction complicated by cardiogenic shock. Eur Heart J 2005.26: 1276–1288.