Secondary Logo

Journal Logo

Adult Circulatory Support

Clinical Outcomes of Impella Microaxial Devices Used to Salvage Cardiogenic Shock as a Bridge to Durable Circulatory Support or Cardiac Transplantation

Cheng, Richard*; Tank, Rikin; Ramzy, Danny; Azarbal, Babak*; Chung, Joshua; Esmailian, Fardad; Kobashigawa, Jon A.*; Moriguchi, Jaime D.*; Arabia, Francisco A.

Author Information
doi: 10.1097/MAT.0000000000000877


Cardiac transplantation remains a definitive therapy for patients with end-stage heart disease. Bridge therapy with durable mechanical circulatory support (MCS) in the form of second-generation left ventricular assist devices and total artificial hearts is increasingly used as a bridge to recovery, transplantation, and/or destination therapy. As we know from the Registry of the International Society for Heart and Lung Transplantation, patients do equally well after transplantation whether they are directly transplanted or bridged with durable MCS with 1 year survival approximating 90%.1 However, outcomes are poor when patients are bridged to transplantation directly from extracorporeal membrane oxygenation (ECMO), or when patients with severe cardiogenic shock are directly implanted with durable MCS.2 Such patients are best stabilized with temporary MCS as part of a strategy to bridge to cardiac transplantation and/or durable MCS. One such device is the Impella device (Abiomed, Danvers, MA), a catheter-based microaxial temporary ventricular assist device that has been used extensively as support for coronary interventions, and more recently to salvage cardiogenic shock. While increasingly used to salvage cardiogenic shock, their use in patients who are otherwise eligible for durable MCS or cardiac transplantation as part of a bridge to advanced therapies strategy is not established.

Extracorporeal membrane oxygenation has been used to support patients in severe cardiogenic shock or arrest before durable MCS or cardiac transplantation, but is associated with worse outcomes with prolonged use, and a myriad of complications3,4 (however with improvements in drivers and implantation techniques contemporary complication rates are likely lower). When used as a direct bridge to transplantation, 1 year survival despite successful transplantation approximates only 50%.1 Other temporary MCS devices are needed.

Impella devices have been used extensively as support for high-risk coronary interventions and for cardiogenic shock complicating acute myocardial infarction. While small randomized trials have demonstrated that their routine use in unselected patients with cardiogenic shock complicating acute myocardial infarction is not warranted,5 their use to salvage and stabilize shock for patients otherwise eligible for durable MCS or cardiac transplantation and as part of a larger advanced therapies strategy has not been studied prospectively. There have been a few small retrospective studies demonstrating that these devices can be used successfully to bridge patients in cardiogenic shock to recovery, expressively to left ventricular assist devices, or directly to heart transplantation,6–8 but the overall experience is small and warrants further investigation.

We therefore sought to describe our experience with the use of Impella to salvage cardiogenic shock in patients who are otherwise eligible for durable MCS or cardiac transplantation as part of a bridge to advance therapies strategy.

Materials and Methods

Patient Cohort and Inclusion Criteria

Procedure charge codes were utilized to identify patients. The medical records of all patients placed on Impella devices between the 50 month period of November 2012 and December 2016 were reviewed. November 2012 was chosen as the starting time point due to the first appearance of Impella CP devices at our institution. Patients in whom Impella was placed for high-risk percutaneous coronary intervention (PCI) or for unplanned hemodynamic deterioration in the cardiac catheterization laboratory, and for whom advanced heart failure and transplantation cardiologists or surgeons were not involved in the care of either before or in a belated fashion in their strategy, were excluded from the analysis. Patients with severe cardiogenic shock bridged with Impella with intent toward durable MCS, heart transplantation or decision were included in the analysis.

The preferred initial device implant was an Impella 5.0 via an axillary cut-down approach. However, if surgical exposure of the subclavian artery revealed a small caliber subclavian artery, an Impella CP may be chosen by the surgeon intraoperatively. Patients are evaluated multiple times a day by the advanced heart failure and transplantation cardiologists and cardiothoracic surgeons. Based on kidney function and liver function studies, lactate, fluid status, hemodynamics, right heart function, and changes in pressor and inotropic requirements, escalation of an Impella CP to an Impella 5.0, the decision to add or replace with an ECMO circuit or escalate to a durable VAD after full committee review is made. In cases of subsequent ECMO placement, Impella devices were routinely left in place as a left ventricular vent.

Impella devices are implanted with the intent of short-term bridge (<4 days for CP and <6 days for 5.0) to recovery or bridge to next therapy whether a durable MCS or heart transplant (OHT). However, clinical circumstances including stability of patient (whether stable on Impella allowing more time to wait for recovery, or not stable enough on Impella to justify implant of a durable MCS) and wait time/availability of organs may lead to a clinical decision to continue support with the existing Impella device due to these unforeseen circumstances.

Baseline demographics including age, sex, number of devices used, median duration of device use, type of Impella device, average duration of Impella-assisted temporary MCS therapy, concomitant ECMO use, Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) profile, recent cardiopulmonary resuscitation (CPR) and concomitant ventilator use are reported. Survival to next therapy (defined as durable MCS or cardiac transplantation), 30 day survival, and 60 day survival are reported. Patients who survived to next therapy are compared with patients who did not. Age, sex, admit source, primary etiology of cardiogenic shock, ischemic cardiomyopathy, coronary artery disease, diabetes mellitus, INTERMACS profile, CPR, concomitant ventilator use, body surface area, body mass index, concomitant ECMO use, type of Impella device, and duration of support are compared. Thirty and 60 day survival conditional on successful bridge to next therapy are reported.

Baseline laboratory values are reported including international normalized ratio (INR), sodium, blood urea nitrogen (BUN), creatinine, albumin, alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin, white blood count, hemoglobin, platelets, and calculated model for end-stage liver disease (MELD) score. Changes in laboratory values between baseline laboratory values and postprocedure day 3 are reported, stratified by patients who survived to next therapy versus patients who did not.

Complications are reported including thrombocytopenia, anemia, vascular site bleeding or complication, any bleeding, lower extremity ischemia, infection, concomitant continuous renal replacement therapy (CRRT), stroke, pump exchange, and pump failure.

Patients who were successfully bridged to cardiac transplantation were compared with patients successfully bridged to durable MCS. Patients suitable for direct bridge to cardiac transplantation were patients who had clinical and laboratory stabilization and recovery on Impella support and with low wait times depending on then listing conditions.

Statistics were performed in utilizing SPSS (IBM, Armonk, NY). Fisher’s exact test was used for two-by-two categorical variables, otherwise χ2 test was utilized. Two-tailed Student’s t-test was used for continuous variables, and a paired samples t-test was used for paired analysis of continuous variables. The Mann–Whitney U test was used for nonparametric variables.


In a 50 month period, 80 Impella devices were utilized in 64 patients with severe cardiogenic shock with intent toward durable MCS or cardiac transplantation at a single institution and with guidance from the institution’s MCS team. Pump failure occurred in four patients requiring Impella exchange. In nine patients, CP devices were replaced with 5.0 devices via a more stable axillary cut-down approach. In one patient, a 5.0 device was replaced with another 5.0 device because of possible contamination and bacteremia after the first device implant. The derivation of this cohort is similar to previously published cohorts and is summarized in Figure 1.4,6–8

Figure 1.
Figure 1.:
Cohort derivation flow chart. ECMO, extracorporeal membrane oxygenation; LV, left ventricle; PCI, percutaneous coronary intervention; Pts, patients; VAD, ventricular assist device.

The average age at implant was 56.2 ± 12.5 years old, and 48/64 (75.0%) were male. INTERMACS profile 1 “crashing and burning” cardiogenic shock was present in 31/64 (48.4%) patients at time of implant, and profile 2 “sliding on inotropes” in the remaining 33/64 (51.6%) patients. One Impella 2.5, 43 Impella CP, and 36 Impella 5.0 devices were used, and the median duration of device use was 7 days (interquartile range: 3–14). Average duration of Impella-assisted temporary MCS therapy was 13.2 ± 15.1 days. Concomitant ECMO occurred in 17/64 (26.7%) of patients. Recent CPR occurred in 17/64 (26.7%) of patients and concomitant ventilator use in 43/64 (67.2%). The most common indication for device implant was acute decompensated heart failure 39/64 (60.9%) with acute myocardial infarction as the second leading indication 22/64 (34.4%).

Survival After Device Implant

Mean clinical follow-up was 232.6 ± 328.3 days. Survival to next therapy occurred in 44/64 (68.8%) of patients. Thirty and 60 day survival were 43/64 (67.2%) and 42/64 (65.6%), respectively. Conditional on patients having survived to next therapy, 30 and 60 day conditional survival were 32/34 (94.1%) and 31/34 (91.2%), respectively. Survivors were less likely to be on renal replacement therapy (15.9%) than those who died (75.0%), p < 0.001, and less likely to be on a ventilator (59.1%) than those who died (85.0%), p = 0.049. Otherwise there were no significant differences between patients who survived and those who did not as summarized in Table 1.

Table 1.
Table 1.:
Baseline Demographic and Clinical Characteristics

Changes in Laboratory Values After Device Implant

Mean baseline laboratory values at time of device implant were consistent with a cardiogenic shock cohort. Average INR (1.5 ± 4.3), BUN (35.1 ± 17.5 mg/dl), creatinine (1.9 ± 1.0 mg/dl), ALT (226.1 ± 582.2 IU/dl), AST (412.6 ± 1142.1 IU/dl), total bilirubin (1.6 ± 1.3 mg/dl), and MELD score (16.1 ± 7.6) were consistent with a cohort of patients in cardiogenic shock with shock liver and acute kidney injury. Survivors were more likely to have a higher albumin (3.7 ± 0.4 vs. 3.3 ± 0.7 g/dl) at baseline than nonsurvivors (p = 0.044). Baseline laboratory values are summarized in Table 1 (Supplemental Digital Content,

In survivors, between baseline and postoperative day (POD) #3 laboratory values, INR significantly improved (1.5 ± 0.3 to 1.3 ± 0.2; p = 0.010). However, bilirubin worsened (1.4 ± 0.9 to 2.5 ± 2.0 mg/dl; p < 0.001), as did MELD score (15.5 ± 7.4 to 17.8 ± 6.5; p = 0.019). Hemoglobin (11.4 ± 2.4 to 8.9 ± 1.5 g/dl; p < 0.001) and platelets (194.1 ± 79.0 to 116.5 ± 53.9 × 109/L; p < 0.001) both worsened. Nominally transaminases improved. Changes in baseline and POD#3 laboratory values in patients who survived are summarized in Table 2 (Supplemental Digital Content,

In nonsurvivors, creatinine worsened (2.2 ± 1.3 to 2.6 ± 1.3 mg/dl; p = 0.050) as did total bilirubin (2.1 ± 2.0 to 5.9 ± 8.5 mg/dl; p = 0.047) and MELD score (18.9 ± 8.0 to 23.0 ± 5.8; p = 0.079). Hemoglobin (11.1 ± 2.5 to 8.5 ± 11.5 g/dl; p = 0.002) and platelets (171.8 ± 77.8 to 93.8 ± 41.2 × 109/L; p = 0.001) both worsened. Changes in baseline and POD#3 laboratory values in patients who died are summarized in Table 3 (Supplemental Digital Content,

By logistic regression, POD#3 total bilirubin was associated with decreased survival to next therapy with an odds ratio of 0.83 (p = 0.099), while POD#3 MELD score had an odds ratio of 0.86 (p = 0.011). For every point increase between baseline and POD#3 total bilirubin, the odds ratio for survival to next therapy was 0.80 (p = 0.111), while every point increase in MELD score had an odds ratio of 0.96 (p = 0.373).


Lower extremity ischemic occurred in 6.3% of the patients, and vascular site bleeding (including documented oozing from the access site) or complication occurred frequently in 21.9% of the cohort. Stroke occurred in 7.8% of the cohort. Pump exchange occurred on 21.9% of the cohort, whether due to pump failure or as part of a strategy to titrate from a CP device to a 5.0 device. Pump failure was less frequent occurring in 6.3% of the cohort. Complications are summarized in Table 4 (Supplemental Digital Content, Complications excluding patients also on ECMO are summarized in Table 2. Patients who survived were less likely to have a stroke (1.3% vs. 20.0%; p = 0.030). Survivors were less likely to be on CRRT (15.9% vs. 75.0%; p < 0.001). For patients who were on CRRT, a logistic regression analysis for days of support and survival to next therapy yielded an odds ratio of 1.01, with a p value of 0.810. The remaining complications were not different between the two groups.

Table 2.
Table 2.:

Direct Bridge to Heart Transplantation

Patients directly bridged to heart transplantation were more likely than patients bridged to durable MCS to have acute decompensated heart failure as their primary etiology of cardiogenic shock (100% vs. 55.6%; p = 0.003), and less likely to have had an ischemic cardiomyopathy (12.5% vs. 77.8%; p < 0.001). They were less sick with less INTERMACS profile 1 cardiogenic shock (12.5% vs. 61.1%; p = 0.005), less CPR (6.3% vs. 33.3%; p = 0.090), and less concomitant ventilator use (31.3% vs. 83.3%; p = 0.004). No patients directly bridged to heart transplantation were on concomitant ECMO (0% vs. 55.6%; p < 0.001). Differences between patients directly bridged to heart transplantation and those bridged to durable MCS are summarized in Table 3. Despite these differences, 30 and 60 day survival condition on surviving to next therapy were similar between the two strategies. Thirty and 60 day conditional survival for patients bridged to durable MCS was 88.9% and 83.3%, respectively. All patients directly bridged to heart transplantation were alive at 30 day, 60 day, and long-term follow-up.

Table 3.
Table 3.:
Heart Transplant as Next Therapy


Impella devices have been used extensively as support for high-risk coronary interventions, and for cardiogenic shock complicating acute myocardial infarction. Their use to salvage and stabilize shock for patients otherwise eligible for durable MCS or cardiac transplantation and as part of a larger advanced therapies strategy has not been studied prospectively. We report on our experience with the use of Impella to salvage cardiogenic shock in patients who are otherwise eligible for durable MCS or cardiac transplantation as part of a bridge to advances therapies strategy and demonstrate survival to next therapy in this largely “crashing and burning” cohort to be a favorable 68.8%. Moreover, in carefully selected patients, the device can be used for as a direct bridge for an average of 21.9 ± 20.8 days to successful heart transplantation in 16 instances with a 30 day, 60 day, and long-term conditional survival of 100%.

There are small cohorts for comparison. In 15 patients receiving temporary devices through an axillary approach long-term survival was 60%.7 In 24 patients bridged with Impella support, survival to next therapy or recovery was 70.8%.6 In the largest single-center comparative cohort, 40 patients were bridged with Impella support over a 6 year period with the intent of bridge to heart transplantation or left ventricular assist device with 75% survival to next therapy, and 30 day survival of 68%.8 Finally, in a recently published three center experience of 58 patients, including patients from the largest single-center comparative cohort, survival to next therapy was 67%.9 These findings are consistent with our findings.

Continuous renal replacement therapy was lower in patients who survived to next therapy but may reflect worsening multisystem organ failure rather than a predictive effect—baseline creatinine was not different in patients who survived and those who did not. Otherwise, we find that baseline demographics and baseline laboratory values at time of implant are poorly associated with successful bridge to next therapy. Moreover, both patients who survived and those who did not saw their total bilirubin increase from implant to POD#3, and a decrease in both hemoglobin and platelets. It is conceivable that hemolysis from the Impella device, whether clinically evident or occult, may be driving these laboratory values, and they should not be used to determine hemodynamic response to mechanical therapy. It is notable however that patients who did not survive had a much larger increase in total bilirubin between the two time points. Moreover, worsening kidney function portended a poor prognosis despite the fact that patients who died were more likely to be on CRRT which may attenuate the observed increase in creatinine.

Complications including anemia and thrombocytopenia are frequent, but pump failure only occurred in 6.3% of patients. Patients who survived to next therapy are less likely to have had vascular site bleeding or complication, stroke, and clinically determined infection. The importance of minimizing vascular site bleeding or complication cannot be understated. Access site complication among other complications are frequently associated with mortality with use of ECMO cannulas.10,11 The use of Impella devices requiring large bore arterial catheters with similar access points would mirror this experience. Despite careful patient selection, vascular site bleeding or complication are frequent in our cohort and ideal access site location (left axillary, right axillary, femoral arteries) and access (surgical cut-down, percutaneous access, and closure) that can minimize vascular site issues remain to be further investigated.

In carefully selected patients, direct bridge to heart transplantation can be safely accomplished. Not surprisingly, we found that patients chosen for this strategy when compared with patients bridged to durable MCS were less sick with less INTERMACS profile 1 cardiogenic shock, less cardiac arrest, and less concomitant ventilator and ECMO use. Nonetheless, these patients can be safely bridged directly to heart transplant with a mean duration of temporary support of over 3 weeks, and many over a month. This strategy may be limited to centers in regions with shorter-than-average anticipated wait times for donor hearts.

Patient selection for advanced therapies is predicated on multiple changing clinical, laboratory, and socioeconomic factors (among others) in the context of a multidisciplinary consulting team. For patients who are foreseeably eligible for advanced therapies but are not currently, temporary MCS allows both stabilization of the clinical situation and an opportunity to evaluate clinical response to determine eligibility. While a competing temporary MCS therapy, ECMO has its associated complications,3 and has not been shown to be a successful direct bridge to heart transplantation with a 1 year survival of 50%.1 The availability of new percutaneous temporary right ventricular assist devices in the Protek Duo (CardiacAssist, Pittsburgh, PA) and Impella RP allows for novel configurations allowing biventricular support traditionally only allowed by ECMO circuits.12 Finally, the selection of ideal left-sided temporary MCS warrants further investigations. While the smaller 14 French Impella CP may be associated with less access site complications and bleeding, factors influencing survival to next therapy, and may be percutaneously inserted even in the axillary position,13,14 the provided support may be inadequate in patients with very severe cardiogenic shock and/or with larger body sizes. While the Impella 5.0 device can be inserted percutaneously in the axillary position,15 this practice is unlikely to become widespread. Moreover, hemolysis is well known to occur with Impella devices, but neither the comparative amount of hemolysis between the CP and 5.0 devices is known nor the clinical relevance of hemolysis.16 Nonetheless, the routine monitoring of lactate dehydrogenase and haptoglobin may be warranted in addition to bilirubin, hemoglobin, and platelets. Further investigations are warranted in determining patient selection and device placement with regards to risk for hemolysis, and whether the HeartMate PHP (St. Jude Medical, St. Paul, MN) may have a role as a temporary MCS system in the future.17 Cardiac enzymes, lactate dehydrogenase, and haptoglobin were not routinely drawn and their association with outcomes cannot be determined. Our institution does not routinely bridge patients from ECMO directly to heart transplantation, but a recent study evaluating the United Network for Organ Sharing (UNOS) database reports a post-transplant survival of 73.1% vs. 93.1% in patients bridged with ECMO versus those bridged with continuous-flow left ventricular assist devices.18 Further studies are warranted to describe the comparative outcomes of ECMO bridge versus Impella bridge.


This study is limited by its retrospective design and differences in survivors and died may represent differences in baseline patient characteristics not fully captured by reported variables. As the size of the study cohort is small performance of multivariate analyses is limited. Eligibility for advanced therapies is complex, multidisciplinary, and multifactorial and cannot be easily captured with traditional clinical variables. As the study is a retrospective analysis, the associated clinical and laboratory findings of survivors are associations, and a prospective multicenter randomized control study would be required to assess the use of these devices as a bridge to eligibility of advances therapies, and to the advances therapies themselves. The approach of a direct bridge to heart transplantation may be limited to regions with shorter-than-average wait times for donor hearts. Our cohort and study were unable to identify a frame of time that could represent a cutoff to consider support futile in patients undergoing CRRT with Impella devices


Impella devices can be used to salvage patients in severe heart failure as a bridge to durable MCS or heart transplantation. Baseline demographics and labs are not predictive of survival. Their use for this indication is increasing and further investigations are warranted to determine the optimal strategy and patient selection.


1. Lund LH, Edwards LB, Dipchand AI, et al.; International Society for Heart and Lung Transplantation: The Registry of the International Society for Heart and Lung Transplantation: Thirty-third Adult Heart Transplantation Report-2016; Focus Theme: Primary Diagnostic Indications for Transplant. J Heart Lung Transplant 2016.35: 1158–1169.
2. 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.
3. Cheng R, Hachamovitch R, Kittleson M, et al. Complications of extracorporeal membrane oxygenation for treatment of cardiogenic shock and cardiac arrest: A meta-analysis of 1,866 adult patients. Ann Thorac Surg 2014.97: 610–616.
4. Cheng R, Ramzy D, Azarbal B, et al. Device strategies for patients in INTERMACS Profiles 1 and 2 Cardiogenic Shock: Double bridge with extracorporeal membrane oxygenation and initial implant of more durable devices. Artif Organs 2017.41: 224–232.
5. Thiele H, Jobs A, Ouweneel DM, et al. Percutaneous short-term active mechanical support devices in cardiogenic shock: A systematic review and collaborative meta-analysis of randomized trials. Eur Heart J 2017.38: 3523–3531.
6. Bansal A, Bhama JK, Patel R, et al. Using the minimally invasive Impella 5.0 via the right subclavian artery cutdown for acute on chronic decompensated heart failure as a bridge to decision. Ochsner J 2016.16: 210–216.
7. Doersch KM, Tong CW, Gongora E, Konda S, Sareyyupoglu B. Temporary left ventricular assist device through an axillary access is a promising approach to improve outcomes in refractory cardiogenic shock patients. ASAIO J 2015.61: 253–258.
8. Lima B, Kale P, Gonzalez-Stawinski GV, Kuiper JJ, Carey S, Hall SA. Effectiveness and safety of the Impella 5.0 as a bridge to cardiac transplantation or durable left ventricular assist device. Am J Cardiol 2016.117: 1622–1628.
9. 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.
10. Sakamoto S, Taniguchi N, Nakajima S, Takahashi A. Extracorporeal life support for cardiogenic shock or cardiac arrest due to acute coronary syndrome. Ann Thorac Surg 2012.94: 1–7.
11. Slottosch I, Liakopoulos O, Kuhn E, et al. Outcomes after peripheral extracorporeal membrane oxygenation therapy for postcardiotomy cardiogenic shock: a single-center experience. J Surg Res 2013.181: e47–e55.
12. Aggarwal V, Einhorn BN, Cohen HA. Current status of percutaneous right ventricular assist devices: First-in-man use of a novel dual lumen cannula. Catheter Cardiovasc Interv 2016.88: 390–396.
13. Mathur M, Hira RS, Smith BM, Lombardi WL, McCabe JM. Fully percutaneous technique for transaxillary implantation of the Impella CP. JACC Cardiovasc Interv 2016.9: 1196–1198.
14. Tayal R, Barvalia M, Rana Z, et al. Totally percutaneous insertion and removal of impella device using axillary artery in the setting of advanced peripheral artery disease. J Invasive Cardiol 2016.28: 374–380.
15. Nakamura K, Krishnan S, Mahr C, McCabe JM. First-in-man percutaneous transaxillary artery placement and removal of the Impella 5.0 mechanical circulatory support device. J Invasive Cardiol 2017.29: E53–E59.
16. Badiye AP, Hernandez GA, Novoa I, Chaparro SV. Incidence of hemolysis in patients with cardiogenic shock treated with impella percutaneous left ventricular assist device. ASAIO J 2016.62: 11–14.
17. Van Mieghem NM, Daemen J, den Uil C, et al. Design and principle of operation of the HeartMate PHPTM (Percutaneous Heart Pump). EuroIntervention 2018.13: 1662–1666.
18. Fukuhara S, Takeda K, Kurlansky PA, Naka Y, Takayama H. Extracorporeal membrane oxygenation as a direct bridge to heart transplantation in adults. J Thorac Cardiovasc Surg 2018.155: 1607–1618.e6.

Impella; severe cardiogenic shock; percutaneous temporary circulatory support; mechanical circulatory support; bridge to decision and transplantation

Supplemental Digital Content

Copyright © 2018 by the ASAIO