Heart support crossroads: aortic Impella 5.5 implantation compared to conventional axillary approach in combined use of extracorporeal membrane oxygenation and Impella (ECMELLA) setting

Introduction

Postcardiotomy shock after cardiac surgery remains challenging, often necessitating advanced mechanical circulatory support. This severe condition, characterized by a sudden and critical decline in cardiac output following surgery, leads to insufficient tissue perfusion and potential organ failure. Many reasons account for the complexity of postcardiotomy shock including pre-existing cardiac conditions, intraoperative complications, and postoperative factors such as inflammation and myocardial stunning (1,2).

Veno-arterial extracorporeal membrane oxygenation (VA-ECMO) is particularly effective in increasing cardiac output and oxygenation post-cardiogenic shock (CS). VA-ECMO ensures adequate tissue perfusion and respiratory support. However, this powerful intervention is not without its drawbacks. One significant concern is that VA-ECMO can increase left ventricular (LV) afterload, which may be detrimental to a recovering heart. The increased afterload can exacerbate myocardial stress and impede recovery in the delicate post-CS phase (3).

The Impella 5.5 device provides enhanced hemodynamic support for patients grappling with severe manifestations of CS. By offering direct LV unloading, the Impella 5.5 reduces the workload on the heart, allowing it to recover more effectively. This unloading effect improves myocardial oxygen supply and decreases oxygen demand, which is crucial for promoting cardiac recovery in patients with postcardiotomy shock. The device’s ability to maintain stable cardiac output while minimizing the stress on the LV makes it a valuable tool in the management of severe cardiac conditions, enhancing the overall recovery potential for these patients (4).

The combined use of Impella with VA-ECMO, generally refered to as combined use of extracorporeal membrane oxygenation and Impella (ECMELLA) leverages the strengths of both devices: VA-ECMO provides robust cardiac and respiratory support, while Impella offers targeted LV unloading (5). This synergy not only supports cardiac function but also enhances overall patient stability, potentially leading to better outcomes in patients experiencing severe CS (6). As interest in this combined therapy grows, so does the need for detailed clinical studies to establish standardized protocols and confirm its long-term benefits. While the axillary artery surgical cut-down is the recommended standard for Impella 5.5 placement, it is not the sole pathway.

The aim of this study was to assess the comparative efficacy and safety of direct aortic cannulation versus axillary artery surgical cut-down for Impella 5.5 implantation, within the ECMELLA setting, for patients with postcardiotomy shock. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-511/rc).

Methods

A retrospective analysis was conducted on patients with postcardiotomy shock at the University Hospital Würzburg. Between January 2023 and August 2024, a total of 24 patients required ECMELLA support for postcardiotomy shock. These patients were divided into two groups: 12 patients underwent axillary artery implantation of the Impella 5.5 device (Abiomed Europe GmbH, Aachen, Germany) forming group 1, while the remaining 12 patients received the device via direct aortic implantation (group 2). The inclusion criteria focused on those who developed severe CS post-cardiac surgery, necessitating advanced mechanical circulatory support. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by institutional Ethics Committee of University Hospital Würzburg (No. 2020050704) and individual consent for this retrospective analysis was waived.

The axillary cannulation was performed using an 8 mm vascular graft sewn end-to-side on the axillary artery for Impella 5.5 implantation, while the central VA-ECMO cannulation was performed as previously described (7). Our novel approach for direct aortic cannulation of Impella device involved a meticulous, dual-purpose surgical technique designed to accommodate the ECMELLA framework. An 8 mm vascular graft (Maquet Hemagard, Getinge Group, Sweden), uniquely modified with a mid-graft opening made by a simple surgical cut, was employed to allow for surgical manipulations. This graft was sewn to the ascending aorta using a side-to-side technique, specifically tailored to integrate both systems for mechanical circulatory support (Figure 1). This configuration facilitated simultaneous implantation: one graft leg was dedicated to the insertion of the Impella 5.5 device and was tunneled suprasternally in the jugular field, while the other was tunneled to the 4th intercostal space for central VA-ECMO arterial cannulation. This suprasternal tunneling into the jugular field was selected to accommodate the higher positioning of the side-to-side graft anastomosis on the distal ascending aorta, which is necessary for optimal Impella 5.5 alignment. Routing the graft in this manner allows the pump to enter the aorta in a straighter, more anatomical path, avoiding tight bends or curves that can complicate insertion and positioning. Importantly, this is performed under direct vision during median sternotomy, minimizing the risk of vascular or nerve injury.

Figure 1 This figure illustrates the anatomical positioning of vascular grafts in a side-to-side manner adjacent to the ascending aorta: (I) the upper limb of the vascular conduit is utilized for Impella implantation. (II) The lower limb of the vascular conduit is utilized for central VA-ECMO cannulation. VA-ECMO, venoarterial extracorporeal membrane oxygenation.

In the case of ECMELLA setting, VA-ECMOs (HLS Module Advanced, Maquet) with Cardiohelp centrifugal pumps (Maquet, Rastatt, Germany) were utilized. To prevent hypothermia, an external heat exchanger (Deltastream HC, Medos) was employed. This dual-cannulation method was developed to optimize hemodynamic support and procedural time, enabling effective management of postcardiotomy shock while allowing chest closure. This approach represents a significant advancement in patient care.

To minimize potential sources of bias, demographic variables, cannulation site, intraoperative parameters, and outcome variables were consistently recorded from objective sources, including medical records, perfusion protocols, the hospital database, and operative reports. This was an all-comers study conducted over a defined time period, reducing selection bias. By including all eligible patients during that timeframe and focusing on standardized data collection methods, we aimed to ensure representative sampling. Primary endpoints focused on comparing clinical risk factors for survival and weaning success in relation to implantation strategy, using consistent criteria across cases to reduce measurement and interpretation bias.

Statistical analysis

Data were analyzed using IBM’s SPSS Version 29.0. Descriptive statistics described population characteristics. Continuous data are reported as means. Nominal data are presented as frequencies. The Shapiro-Wilk test checked normal distribution. The t-test analyzed normal data; the Mann-Whitney U test analyzed non-normal data. Fisher’s exact test and Chi-squared test were used for nominal data. Significance was set at P<0.05, with 95% confidence intervals calculated for accuracy.

Results

There were no significant differences in preoperative clinical scores or demographics between the two groups (Table 1). The majority of cases required an emergency procedure. Both the axillary and aortic implantation groups exhibited comparable baseline characteristics, indicating a homogenous patient population for evaluating the efficacy and safety of the two surgical approaches. The index surgical procedures included isolated coronary artery bypass grafting (CABG) (n=10), isolated aortic valve replacement (AVR) (n=1), combined CABG + AVR (n=5), aortic valve (AV) valve surgery (n=6), and more complex combinations such as CABG or AVR with mitral valve repair (n=1), and one case of isolated aortic surgery (n=1). The distribution of procedures is detailed in Table 1.

Both implantation techniques were technically successful and allowed for adequate hemodynamic support in patients with postcardiotomy shock. Postoperative assessments revealed that both groups achieved significant improvements in cardiac output and overall hemodynamic stability. Importantly, no significant differences were observed in cannulation-related complication rates between the two groups post-operatively (Table 2). These findings suggest that both the axillary and aortic implantation techniques are feasible and safe approaches for Impella 5.5 insertion in the ECMELLA setting, even in this critically ill population. It is important to note, however, that the high in-hospital mortality reflects the severity and complexity of postcardiotomy shock, rather than the limitations of either implantation technique.

The length of stay in the intensive care unit (ICU) was comparable between the two groups, with 14.5±13.5 days for group 2 patients versus 16.50±13.2 days for group 1 patients (P=0.72). Additionally, the usage of levosimendan was similar in both groups. Successful weaning from mechanical circulatory support was achieved in 5 of 12 patients (41.7%) in the aortic group and in 4 of 12 patients (33.3%) in the axillary group. The mean duration of mechanical circulatory support was similar between groups: 11.08±7.78 days in the aortic group and 8.17±6.93 days in the axillary group.

The rates of re-exploration for bleeding were similar, with 58.3% in the aortic group and 41.7% in the axillary group (P=0.41). Nevertheless, the incidence of cannulation site bleeding was very low in both groups (0% vs. 8.3%, P=0.31). Temporary dialysis was required in 41.7% of patients in the aortic group compared to 50.0% in the axillary group (P=0.19).

Notably, cardiopulmonary bypass (CPB) time was significantly longer in the axillary group (247.8±145.3 minutes) compared to the aortic group (137.3±77.2 minutes, P=0.03), indicating a more complex procedure for instituting ECMELLA support in the axillary group. Median cross-clamp times were comparable between the axillary and aortic groups, reflecting similar underlying surgical complexity.

In-hospital mortality rates were comparable between the groups, with 66.7% in group 2 and 75.0% in group 1 (P=0.65) (Table 2). The number of red blood cell transfusions was 28.0±15.0 unit in group 1 versus 23.6±14.8 unit in group 2 (P=0.48). Platelet transfusions were 8.0±5.8 unit in group 1 versus 6.3±3.3 unit in group 2 (P=0.40). A total of 17 patients died during the in-hospital period. The most common causes of death were multiorgan dysfunction syndrome (MODS; n=6), cerebrovascular events (stroke, n=4), and low cardiac output syndrome (LCOS; n=3). Additional causes included septic shock (n=3) and mesenteric ischemia (n=1). A breakdown by cannulation group is provided in Table 3.

Discussion

We showed that direct aortic cannulation for Impella 5.5 as part of the ECMELLA concept can be performed safely and accurately. This technique involves using a single vascular graft to anchor both devices, the Impella 5.5 and arterial ECMO canula, and provides a stable and secure point of access. Direct aortic cannulation for VA-ECMO can potentially reduce the risks associated with femoral access, such as bleeding and limb ischemia, and simplify the overall management of ECMELLA. By centralizing the insertion site, this approach may enhance procedural efficiency, expedite procedural time and thereby improve patient outcomes, making it a valuable option in the advanced management of CS.

In our previous study, we demonstrated that central VA-ECMO cannulation offers several advantages in managing postcardiotomy shock, including improved survival rates without increasing the need for surgical revisions (7). Central VA-ECMO also offers notable hemodynamic advantages over peripheral ECMO. A study by Gu et al. demonstrated that peripheral ECMO was associated with lower perfusion to the lower limbs, higher wall shear stress at the femoral artery bifurcation, and increased oscillatory shear index in the aortic arch and femoral access sites, all factors that increase the risk of vascular injury and limb ischemia. In contrast, central ECMO showed improved hemodynamic energy and better preservation of pulsatile flow. These physiological benefits may translate into reduced rates of hemolysis, oxygenator thrombosis, and peripheral vascular complications, making central access a valuable alternative in complex or prolonged support cases (8). Despite its advantages in providing high extracorporeal blood flow and sufficient gas exchange, VA-ECMO has significant disadvantages, particularly the continuous nonphysiological infusion of blood into the arterial vasculature, especially in cases of peripheral ECMO cannulation with retrograde flow. In postcardiotomy shock, the LV lacks preload and contractile reserve to overcome the increased afterload, leading to increased myocardial oxygen demand on a failing ventricle. Therefore, LV unloading with VA-ECMO reduces myocardial oxygen consumption, which addresses manifested complications and promotes LV recovery (9).

Early active LV unloading after VA-ECMO implantation is associated with a lower mortality risk compared to delayed LV unloading, and is not linked to an increased likelihood of complications. Avoiding prolonged periods of LV overload caused by retrograde aortic perfusion may help explain this association. Findings from other studies underscore the importance of LV unloading in managing CS and suggest that early active LV unloading may improve outcomes in CS patients treated with VA-ECMO (10). Moreover, Shrage et al. in recent multi-centric study observed an increasing mortality risk with every extra hour of delay of active LV unloading (11).

The concept of establishing ECMELLA using a single arterial access has been previously described, where both the arterial cannula and Impella 5.5 can be successfully inserted using the same vascular graft in the axillary artery. This technique has proven to be very useful in cases of CS (12,13). Our approach represents a significant step forward in the treatment of severe postcardiotomy shock. In scenarios where the aorta is already exposed, further dissection to expose the axillary artery can pose additional bleeding risks and is time-consuming. By eliminating the need for this additional step, our method reduces complications and allows for faster institution of active LV unloading offering a new paradigm in the use of advanced mechanical support devices The longer CPB times observed in the axillary group were primarily due to the need for additional surgical steps during device implantation, including exposure of the axillary artery, tunneling, and securing the vascular graft. In many cases, intraoperative imaging with a C-arm fluoroscopy was required to guide accurate placement of the Impella 5.5, further extending procedural time. This was not routinely necessary in the aortic group. Additionally, in several patients, CPB had to be re-initiated or prolonged after an initial weaning attempt to safely complete the axillary implantation. These factors likely contributed to the increased CPB duration observed in the axillary group compared to the more direct and fluoroscopy-free aortic approach. When placing the Impella 5.5 using our method, it is crucial to ensure that the insertion graft leg has an optimal length to avoid stasis and clotting. The Impella can be implanted directly under transesophageal control without the need for fluoroscopy. The other leg of the vascular graft can be tunneled through the intercostal space, enabling direct central cannulation. From our experience, performing the side-to-side anastomosis relatively distal on the ascending aorta is crucial to provide ample space for Impella implantation.

This approach allows for chest closure and de-escalation of mechanical circulatory support without reopening the chest. After initial stabilization of the patient, and as right ventricular function and oxygenation improve, it is possible to wean and explant the VA-ECMO as bedside procedure under local anesthesia. Subsequently, the same approach can be applied to explantation of the Impella when support is no longer needed. This reduces complications associated with VA-ECMO, such as hemorrhage and limb ischemia, which can be fatal (14). Additionally, this technique eliminates the need for extra vascular loops and mitigates the danger of hyperperfusion of the arm associated with axillary cannulation.

This study has several limitations. First, the retrospective design may introduce selection and reporting biases. Second, the sample size was relatively small, limiting the statistical power to detect differences in rare outcomes. Additionally, this was a single-center study, which may affect the generalizability of the findings to other institutions with different protocols and expertise.

Conclusions

Both the axillary and direct aortic cannulation techniques for Impella 5.5 implantation in the ECMELLA setting were feasible and safe, providing viable options for achieving hemodynamic stabilization in patients with postcardiotomy shock. Although the axillary approach was associated with longer CPB times, both techniques demonstrated similar rates of in-hospital mortality and ICU stays—reflecting the severity of the underlying condition rather than differences in cannulation strategy. The direct aortic cannulation approach offers a significant advantage with shorter implantation times, enhancing procedural efficiency and maintaining safety. This technique allows for chest closure and reduced CPB time, making it a practical and beneficial option in various clinical scenarios.

Acknowledgments

We would like to express deep gratitude to Kiril Penov, who made significant contributions to this research. His expertise, dedication, and passion were invaluable to the progress of this work. Sadly, Kiril Penov passed away during the preparation of this manuscript. He was not only a respected colleague, but also a dear friend.

The graphic design in this manuscript was provided by Motion & Strategy, Virtual Reality, Film, 3D.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-511/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-511/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-511/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-511/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by institutional Ethics Committee of University Hospital Würzburg (No. 2020050704) and individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Pérez Vela JL, Llanos Jorge C, Duerto Álvarez J, et al. Clinical management of postcardiotomy shock in adults. Med Intensiva (Engl Ed) 2022;46:312-25. [Crossref] [PubMed]
2. Corujo Rodriguez A, Richter E, Ibekwe SO, et al. Postcardiotomy Shock Syndrome: A Narrative Review of Perioperative Diagnosis and Management. J Cardiothorac Vasc Anesth 2023;37:2621-33. [Crossref] [PubMed]
3. Ezad SM, Ryan M, Donker DW, et al. Unloading the Left Ventricle in Venoarterial ECMO: In Whom, When, and How? Circulation 2023;147:1237-50. [Crossref] [PubMed]
4. Sugimura Y, Bauer S, Immohr MB, et al. Clinical outcomes of hundred large Impella implantations in cardiogenic shock patients based on individual clinical scenarios. Artif Organs 2023;47:1874-84. [Crossref] [PubMed]
5. Fiorelli F, Panoulas V. Impella as unloading strategy during VA-ECMO: systematic review and meta-analysis. Rev Cardiovasc Med 2021;22:1503-11. [Crossref] [PubMed]
6. Nersesian G, Potapov E. Mechanical circulatory support with ECMELLA approach in severe cardiogenic shock patients-state of the art. J Thorac Dis 2023;15:4555-7. [Crossref] [PubMed]
7. Radakovic D, Hamouda K, Penov K, et al. Central Versus Peripheral Arterial Cannulation for Veno-Arterial Extracorporeal Membrane Oxygenation in Post-Cardiotomy Patients. ASAIO J 2021;67:67-73. [Crossref] [PubMed]
8. Gu K, Zhang Y, Gao B, et al. Hemodynamic Differences Between Central ECMO and Peripheral ECMO: A Primary CFD Study. Med Sci Monit 2016;22:717-26. [Crossref] [PubMed]
9. Uriel N, Sayer G, Annamalai S, et al. Mechanical Unloading in Heart Failure. J Am Coll Cardiol 2018;72:569-80. [Crossref] [PubMed]
10. Radakovic D, Zittermann A, Rojas SV, et al. Left Ventricular Unloading in Patients on Venoarterial Extracorporeal Membrane Oxygenation Therapy in Cardiogenic Shock: Prophylactic Versus Bail-Out Strategy. Life (Basel) 2023;13:582. [Crossref] [PubMed]
11. Schrage B, Sundermeyer J, Blankenberg S, et al. Timing of Active Left Ventricular Unloading in Patients on Venoarterial Extracorporeal Membrane Oxygenation Therapy. JACC Heart Fail 2023;11:321-30. [Crossref] [PubMed]
12. Eulert-Grehn JJ, Starck C, Kempfert J, et al. ECMELLA 2.0: Single Arterial Access Technique for a Staged Approach in Cardiogenic Shock. Ann Thorac Surg 2021;111:e135-7. [Crossref] [PubMed]
13. Nersesian G, Lewin D, Ott S, et al. Temporary extracorporeal life support: single-centre experience with a new concept. Interdiscip Cardiovasc Thorac Surg 2024;38:ivae043. [Crossref] [PubMed]
14. Truby LK, Takeda K, Mauro C, et al. Incidence and Implications of Left Ventricular Distention During Venoarterial Extracorporeal Membrane Oxygenation Support. ASAIO J 2017;63:257-65. [Crossref] [PubMed]