Secondary Logo

Journal Logo

Effect of Impella During Veno-Arterial Extracorporeal Membrane Oxygenation on Pulmonary Artery Flow as Assessed by End-Tidal Carbon Dioxide

Eliet, Jacob*; Gaudard, Philippe*,†; Zeroual, Norddine*; Rouvière, Philippe; Albat, Bernard; Mourad, Marc*; Colson, Pascal H.*,§

doi: 10.1097/MAT.0000000000000662
Adult Circulatory Support

Peripheral veno-arterial extracorporeal membrane oxygenation (VA ECMO) exposes the patient to a pulmonary blood flow bypass and a left ventricle afterload increase. Impella, a catheter-mounted microaxial rotary pump, has been proposed for left ventricle (LV) unloading in combination with VA ECMO. In order to assess the effect of Impella on pulmonary flow and LV preload, we checked Doppler pulmonary artery velocity–time integral (pVTI) and LV diastolic diameter (LVED) by transesophageal echocardiography and end-tidal carbon dioxide (EtCO2) during a step-by-step increase in Impella flow (Impella ramp test). From 134 patients on VA ECMO retrieved from our database, 27 (20%) have benefited secondary Impella implantation, out of which 11 patients had available EtCO2, pVTI, and LVED measurements at various levels of Impella speeds. We observed a proportional increases in pVTI and EtCO2 and decrease in LVED (p ≤ 0.001) during Impella flow increase. There was a significant correlation between EtCO2 and pVTI (Pearson correlation coefficient 0.64; p = 0.006). The study shows that Impella improves pulmonary flow, an effect that can be easily measured by EtCO2 monitoring, and ensures LV discharge, allowing adapting Impella flow adequately to patient’s individual needs.

From the *Department of Anesthesiology and Critical Care Medicine, Arnaud de Villeneuve Academic Hospital, F-34295 Montpellier, France

PhyMedExp, University of Montpellier, INSERM U1046, Montpellier, France

Department of Cardiac Surgery, Arnaud de Villeneuve Academic Hospital, F-34295 Montpellier, France

§Institut de Génomique Fonctionnelle, Department of Endocrinology, CNRS UMR 5203, INSERM U1191, University of Montpellier, Montpellier, France.

Submitted for consideration March 2017; accepted for publication in revised form August 2017.

Philippe Gaudard contributed equally to this article.

Disclosure: The authors have no conflicts of interest to report.

Correspondence: Pascal Colson, Département d’Anesthésie Réanimation Arnaud de Villeneuve, CHRU Montpellier, 371 avenue Doyen Giraud, F-34295 Montpellier, France. Email:

Transient circulatory support (TCS) devices are increasingly used in cardiogenic shock (CS) and recommended in guidelines of various scientific societies.1,2 Peripheral veno-arterial extracorporeal membrane oxygenation (VA ECMO) ensures circulatory support, meaning peripheral circulation and organ function, expecting myocardial recovery (bridge to recovery) or other therapeutic options (bridge to cardiac transplantation or left ventricle [LV] assistance).3–5

Veno-arterial extracorporeal membrane oxygenation is the most frequently used device for TCS as it is a quick, safe, and cost-effective option.3,6 However, VA ECMO exposes the patient to a pulmonary blood flow bypass and an increase in LV afterload. Both may result in pulmonary edema, LV distension, and left heart cavities or pulmonary artery thrombosis. Fifteen percent to 20% patients under VA ECMO may need LV unloading by alternative techniques.7–12 Several strategies have been considered: either through creating artificial left-to-right shunt using blade or balloon atrial septotomy7–9 or direct left heart unloading with surgical left atrial or ventricle cannulation8,9 or peripheral transeptal LV or left atrial cannulation,10,11 while other authors suggest to use intra-aortic balloon pump.12

Impella (Abiomed Europe GmbH, Aachen, Germany), a catheter-mounted microaxial rotary pump, is an another option. It has already been used for LV unloading in combination with VA ECMO and its efficacy reported in case report13 or case series.14–16 Impella placed in the LV across the aortic valve, unloads the LV, reduces LV wall stress, and improves pulmonary blood flow. Moreover, as opposed to the other LV unloading alternatives, Impella offers the opportunity to adjust flow and consequently the LV discharge.

Impella has become the technique of reference for LV unloading during VA ECMO support in our unit, since 2009.16 With improved experience, it became obvious that insertion and implementation of the device could be conducted according to a specific protocol aimed at adjusting the Impella speed flow to get an optimal LV unloading without suctioning. Moreover, we observed that pulmonary flow can be assessed by end-tidal carbon dioxide (EtCO2) under VA ECMO, a data that have not been described during TCS yet. Therefore, we anticipated that EtCO2 monitoring may be used during VA ECMO, eventually to determine effective Impella flow.

The aim of this report is to describe the effect of Impella flow increase on pulmonary flow and LV preload from a cohort of patients admitted in our intensive care unit (ICU) for VA ECMO who required LV unloading.

Back to Top | Article Outline

Patients and Methods


We retrospectively retrieved from the TCS database of our 14-bed ICU from April 2009 to December 2013 patients admitted for CS. Patients treated with VA ECMO, who benefited from Impella implantation to unload the LV, were studied.

The study was approved by the Montpellier University Hospital institutional review board, which, because of the retrospective nature of the study, waived the need for informed consent.

Back to Top | Article Outline

Transient Circulatory Support Devices and Implantation

All devices were implanted by an experienced cardiac surgeon.

Veno-arterial extracorporeal membrane oxygenation was implanted in the operating room, in ICU, or catheterization laboratory according to the emergency and available staff. It consisted of polyvinyl chloride tubing with a membrane oxygenator (PH.I.S.I.O and EOS; Sorin Group, Clamart, France), a centrifugal pump (Stockert; Sorin Group), and percutaneous or surgically inserted arterial and venous femoral cannulae (Fem-Flex and Fem-Track; Edwards Lifesciences, Guyancourt, France) with an additional 7 Fr cannula inserted distally into the femoral artery to prevent lower limb ischemia. The lower pump rate was set to provide flow necessary for adequate tissue perfusion (mainly, normal central venous oxygen hemoglobin saturation (SvcO2) and blood lactate level).

Impella implantation during VA ECMO support was decided according to criteria selected by our heart team, including severe LV overload with severe pulmonary edema, heavy spontaneous contrast in the left heart cavities on echocardiography, or loss of LV ejection (from aortic velocity-–time integral < 5 cm or pulse pressure < 10 mm Hg including loss of aortic valve opening)

The Impella devices (Abiomed Europe GmbH) consisted in either 5.0 catheter or a 2.5 catheter. The Impella 5.0 is a 9 Fr catheter-mounted microaxial intracardiac pump (21 Fr) which needed surgically insertion through the femoral artery (arteriotomy) or the right axillary artery (through a vascular graft) in the operating room or in the cardiac catheterization laboratory. The Impella 2.5 is a 12 Fr pigtail catheter-based microaxial flow device (catheter:4 mm, 12 Fr outer diameter), inserted retrogradely into the LV across the aortic valve via the femoral artery through a 13 Fr sheath. Impella position across the aortic valve into the LV was controlled using fluoroscopy or transesophageal echocardiography.

Impella flow is determined by nine various support levels (P1 to P9) according to the device design, which correspond to incremental speed rotation levels. Impella 2.5 and Impella 5.0 provide continuous blood flow by transvalvular active support (flow estimates respectively up to 2.5 L/min and 5.0 L/min), with direct LV unloading. Since January 2011, a protocol of Impella settings has been designed to titrate LV unloading. Basically, after Impella implantation and its correct position checked, the protocol consisted in an increase in Impella flow under transesophageal echocardiography control. An incremental increase of speed flow, starting at the lowest speed (P1), was followed by a step-by-step increase, every 5 minutes (Impella ramp test).

Back to Top | Article Outline

Impella Effect Assessment

End-tidal carbon dioxide reflects venous return to the heart and parallels pulmonary flow for a given ventilation17,18; therefore, EtCO2 was recorded during Impella settings.

At the same times, transesophageal echocardiography allowed pulmonary artery flow measurement by Doppler (pulmonary artery velocity–time integral, pVTI) and LV preload was assessed by LV size, namely LV endiastolic diameter (LVED).

Back to Top | Article Outline

Data Collection

Demographic data and TCS characteristics were collected: age, sex, body mass index, etiology of CS, Impella model and implantation site, cardiac arrest before TCS, time from VA ECMO implantation to Impella implantation, and Simplified Acute Physiology Score (SAPS) II (20).

The following clinical variables were collected at Impella implantation: Sepsis-related Organ Failure Assessment (SOFA) score (21), mean arterial pressure (MAP), invasive mechanical ventilation (MV) support, inotropic and vasopressor support (dobutamine, epinephrine, or norepinephrine infusion), and blood lactate.

The reasons to unload the LV were noted.

During Impella titration protocol, heart rate, MAP, pulse pressure (systolic pressure − diastolic pressure), Impella and VA ECMO flows, LVED, Doppler aortic velocity–time integral (AoVTI), pVTI, EtCO2, and carbon dioxide arterial pressure (PaCO2) have been collected at the following Impella times: P1, P3, P6, Pm (maximum speed reached during the protocol), and Pc (level of speed chosen by the clinician).

Back to Top | Article Outline

Statistical analysis

All data are presented as absolute values and percentages (%) for categorical variables or median and interquartile range (IQR) for continuous data. Paired Wilcoxon and Friedman test was used to compare variables during the Impella titration protocol. The relationship between pVTI and EtCO2 was analyzed by linear regression analysis and assessed by Pearson product-moment correlation coefficient.

Statistical significance was defined as p value < 0.05. Analyses were performed using XLSTAT 2013 software (Addinsoft, New York, NY).

Back to Top | Article Outline


Population Studied

From 134 patients on VA ECMO for CS retrieved from the database, 27 (20%) have benefited secondary Impella implantation to unload the LV, and 11 patients had measurements of pulmonary artery flow and LV size during Impella settings (Figure 1).

Figure 1

Figure 1

Patient baseline characteristics are summarized in Table 1.

Table 1

Table 1

Back to Top | Article Outline

Impella Implantation

Indication of unloading LV was mostly a combination of several criteria: severe pulmonary edema in 12 (44%) patients, spontaneous echocontrast in left heart cavities in 12 (44%) patients, or loss of LV ejection in 22 (81%) patients including nine (33%) patients with no aortic valve opening.

Back to Top | Article Outline

Impella Effects

Hemodynamic changes according to stepwise increment of Impella flows are reported in Table 2. VA ECMO flow (4.2 [3.2–4.6] L/min) and ventilation settings were kept constant which resulted in a PaCO2 of 38.5 (35.7–40) mm Hg before Impella implantation.

Table 2

Table 2

Impella implantation did not change blood gas values: oxygen arterial pressure (PaO2) from 113 (99–140) to 122 (99.5–153) and PaCO2 was 38 (35–41.5) mm Hg after Impella settings.

Impella flows increased with program values and resulted in a proportional increase in pVTI and EtCO2 and decrease in LVED (Table 2). Pc value was 6 (4–8), lower than the Pm 8 (6–9) to avoid excessive suctioning but was associated with a significant increase in MAP, pVTI, and EtCO2 and a significant decrease in LVED from P1 (Table 2). There was a significant correlation between EtCO2 and pVTI (Figure 2). However, individual variations in LVED and EtCO2 during Impella settings are quite different from one patient to another (Figure 3). Of note, four of the 11 patients had EtCO2 ≤ 5 mm Hg before Impella setting (Figure 2).

Figure 2

Figure 2

Figure 3

Figure 3

Back to Top | Article Outline


For patients on VA ECMO for CS who present very low native cardiac output or severe LV overload, Impella notably decreased LV overload and restored pulmonary flow. The study shows that Impella speed settings could be done using EtCO2 monitoring, possibly combined with echocardiography.

Among various solutions to discharge LV during VA ECMO, invasive or less invasive techniques using percutaneous approaches create a de novo anatomical shunt between left and right heart cavities, which may unload the LV but may not restore pulmonary blood flow.7–9,11,19,20 Alternatively, Petroni et al.12 proposed to use intra-aortic balloon pump. Indeed, the flow coming from the VA ECMO into the aorta backward to the LV is stopped during balloon inflation, which can represent up to 60% of the time. Therefore, LV residual ejection may decrease LVED and pulmonary artery occlusion pressure but at the expense of reduced circulatory support beyond the inflated balloon and without evidence of pulmonary blood flow restoration.21

Impella device is able to unload left heart cavities as well as restoring pulmonary blood flow during VA ECMO and may have some outstanding advantages. It has already been reported in case report or case series,13–15 and quality of left heart cavities decompression was confirmed repeatedly.22–24 In animal models of myocardial infarction, Impella reduced myocardial oxygen consumption and infarct size22 or conserved calcium cycling and improved heart function.24 When compared with intra-aortic balloon pump in an animal model of severe LV failure (acute mitral regurgitation), Impella is capable of more effective cardiac unloading and circulatory support.23 In our study, LV unloading is confirmed by a substantial decrease of the LVED from 49 (42–57) mm to 30 (21–45) mm.

At the same time, pulmonary blood flow increased significantly and gradually with Impella flow increase. Pulmonary flow increase was assessed by Doppler measurement of pulmonary flow and EtCO2. Of note, most patients have no or very low transpulmonary flow, lower than the accepted limit of the pulmonary catheter monitoring technique (i.e., 2 L/min),25 which makes it therefore irrelevant in this clinical condition. Moreover, possible loss of indicator injection in right atrium because of venous drainage by the VA ECMO machine prevents from appropriate measurement of cardiac output by pulmonary thermodilution, as part of the flow is derived to the VA ECMO circuit. Conversely, EtCO2 that reflects venous return and pulmonary flow provided ventilation is unchanged17,18 seems an attractive technique to assess pulmonary blood flow restoration and evolution. Previous studies have already demonstrated that the increase in pulmonary blood flow may be easily appreciated by EtCO2.26,27 In our study, EtCO2 increases parallel to pVTI, which is a less accessible measure, so that EtCO2 could be a continuous and very convenient surrogate to appreciate pulmonary flow during VA ECMO.

Furthermore, in contrast to most other invasive techniques aimed at reducing pressure in the left cavities, Impella effect can be adjusted by adapting flow speed to LV size (titration). We observed that increasing gradually the Impella speed resulted in gradual EtCO2 increase and LV discharge. Both parameters can be used to set the appropriate Impella speed, without exposing the patient to a risk of cavitation in the LV because of excessive suctioning, which may aggravate hemolysis. Besides the observed interindividual variability of the response on EtCO2 or LVED to the increase in Impella flow, it is noticeable that both variations plateaued before maximal speed in most patients although at various times. Therefore, Impella speed can be tailored to the patient’s need (Figure 4).

Figure 4

Figure 4

Our study has several limitations. First, the retrospective, single-center nature and the sample size of this study are important limitations. However, selection of patients receiving VA ECMO in severe CS requiring LV unloading with Impella provides an original contribution to assess how to manage the tuning of TCS combination when VA ECMO and Impella are associated.

In conclusion, Impella effect on pulmonary artery flow can be assessed by EtCO2, which could be useful for Impella flow setting, to allow LV unloading while avoiding excessive suctioning.

Back to Top | Article Outline


1. McMurray JJ, Adamopoulos S, Anker SD, et alESC Committee for Practice Guidelines: ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The task force for the diagnosis and treatment of acute and chronic heart failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J 2012.33: 1787–1847,
2. Yancy CW, Jessup M, Bozkurt B, et al2013 ACCF/AHA guideline for the management of heart failure: A report of the American College of Cardiology Foundation/American Heart Association Task Force on practice guidelines Circulation 2013128: e240–327,
3. O’Gara PT, Kushner FG, Ascheim GD, et alAmerican College of Emergency Physicians, Society for Cardiovascular Angiography and Interventions: 2013 ACCF/AHA guideline for the management of ST-elevation myocardial infarction: A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines J Am Coll Cardiol 201361: e78–e140,
4. Windecker S, Kolh P, et alAuthors/Task Force members, 2014 ESC/EACTS Guidelines on myocardial revascularization: The Task Force on Myocardial Revascularization of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS)Developed with the special contribution of the European Association of Percutaneous Cardiovascular Interventions (EAPCI) Eur Heart J 201435: 2541–2619,
5. Steg PG, James SK, Atar D, et alTask Force on the management of ST-segment elevation acute myocardial infarction of the European Society of Cardiology (ESC): ESC Guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation Eur Heart J 201233: 2569–2619,
6. Jaski BE, Ortiz B, Alla KR, et alA 20-year experience with urgent percutaneous cardiopulmonary bypass for salvage of potential survivors of refractory cardiovascular collapse. J Thorac Cardiovasc Surg 2010.139: 753–757.e1e2,
7. Koenig PR, Ralston MA, Kimball TR, Meyer RA, Daniels SR, Schwartz DCBalloon atrial septostomy for left ventricular decompression in patients receiving extracorporeal membrane oxygenation for myocardial failure. J Pediatr 1993.122: S95–S99,
8. Hacking DF, Best D, d’Udekem Y, et alElective decompression of the left ventricle in pediatric patients may reduce the duration of venoarterial extracorporeal membrane oxygenation. Artif Organs 2015.39: 319–326,
9. Kotani Y, Chetan D, Rodrigues W, et alLeft atrial decompression during venoarterial extracorporeal membrane oxygenation for left ventricular failure in children: current strategy and clinical outcomes Artif Organs 201337: 29–36,
10. Alkhouli M, Narins CR, Lehoux J, Knight PA, Waits B, Ling FSPercutaneous decompression of the left ventricle in cardiogenic shock patients on venoarterial extracorporeal membrane oxygenation. J Card Surg 2016.31: 177–182,
11. Aiyagari RM, Rocchini AP, Remenapp RT, Graziano JNDecompression of the left atrium during extracorporeal membrane oxygenation using a transseptal cannula incorporated into the circuit. Crit Care Med 2006.34: 2603–2606,
12. Petroni T, Harrois A, Amour J, et alIntra-aortic balloon pump effects on macrocirculation and microcirculation in cardiogenic shock patients supported by venoarterial extracorporeal membrane oxygenation. Crit Care Med 201442: 2075–2082,
13. Vlasselaers D, Desmet M, Desmet L, Meyns B, Dens JVentricular unloading with a miniature axial flow pump in combination with extracorporeal membrane oxygenation. Intensive Care Med 2006.32: 329–333,
14. Cheng A, Swartz MF, Massey HTImpella to unload the left ventricle during peripheral extracorporeal membrane oxygenation. Asaio J Am Soc Artif Intern Organs 1992 2013.59: 533–536,
15. Koeckert MS, Jorde UP, Naka Y, Moses JW, Takayama HImpella LP 2.5 for left ventricular unloading during venoarterial extracorporeal membrane oxygenation support. J Card Surg 2011.26: 666–668,
16. Gaudard P, Mourad M, Eliet J, et alManagement and outcome of patients supported with Impella 5.0 for refractory cardiogenic shock. Crit Care 2015.19: 363,
17. Maslow A, Stearns G, Bert A, et alMonitoring end-tidal carbon dioxide during weaning from cardiopulmonary bypass in patients without significant lung disease. Anesth Analg 2001.92: 306–313,
18. Shibutani K, Muraoka M, Shirasaki S, Kubal K, Sanchala VT, Gupte PDo changes in end-tidal PCO2 quantitatively reflect changes in cardiac output? Anesth Analg 1994.79: 829–833,
19. Johnston TA, Jaggers J, McGovern JJ, O’Laughlin MPBedside transseptal balloon dilation atrial septostomy for decompression of the left heart during extracorporeal membrane oxygenation. Catheter Cardiovasc Interv 1999.46: 197–199,
20. Cheung MM, Goldman AP, Shekerdemian LS, Brown KL, Cohen GA, Redington ANPercutaneous left ventricular “vent” insertion for left heart decompression during extracorporeal membrane oxygenation. Pediatr Crit Care Med 2003.4: 447–449,
21. Sauren LDC, Reesink KD, Selder JL, Beghi C, van der Veen FH, Maessen JGThe acute effect of intra-aortic balloon counterpulsation during extracorporeal life support: An experimental study Artif Organs 200731: 31–38,
22. Meyns B, Stolinski J, Leunens V, Verbeken E, Flameng WLeft ventricular support by catheter-mounted axial flow pump reduces infarct size. J Am Coll Cardiol 2003.41: 1087–1095,
23. Reesink KD, Dekker AL, Van Ommen V, et alMiniature intracardiac assist device provides more effective cardiac unloading and circulatory support during severe left heart failure than intraaortic balloon pumping. Chest 2004.126: 896–902,
24. Wei X, Li T, Hagen B, et alShort-term mechanical unloading with left ventricular assist devices after acute myocardial infarction conserves calcium cycling and improves heart function. JACC Cardiovasc Interv 2013.6: 406–415,
25. Yelderman MContinuous measurement of cardiac output with the use of stochastic system identification techniques. J Clin Monit 1990.6: 322–332,
26. Gazmuri RJ, Weil MH, Bisera J, Rackow ECEnd-tidal carbon dioxide tension as a monitor of native blood flow during resuscitation by extracorporeal circulation. J Thorac Cardiovasc Surg 1991.101: 984–988,
27. Jin X, Weil MH, Tang W, et alEnd-tidal carbon dioxide as a noninvasive indicator of cardiac index during circulatory shock. Crit Care Med 2000.28: 2415–2419,

extracorporeal membrane oxygenation; heart failure; circulatory support devices; Impella

Copyright © 2018 by the American Society for Artificial Internal Organs