Journal Logo


Venoarterial Extracorporeal Membrane Oxygenation in Elderly Patients With Refractory Cardiogenic Shock: Is Age Truly Just a Number?*

Jabaley, Craig S. MD1,2

Author Information
doi: 10.1097/CCM.0000000000004716
  • Free
  • Editor's Choice

The clinical utilization of extracorporeal membrane oxygenation (ECMO) has seen remarkable expansion since its inception in the early 1970s, including a period of exponential growth in the United States (1). Venoarterial (VA) ECMO—in which venous blood return is diverted from the heart extracorporeally, oxygenated, and returned under pressure to the systemic arterial circulation—can temporize even severe forms of cardiopulmonary failure. Procedural, clinical, and technical advances have served to make VA ECMO more accessible, convenient, and safe than ever. Indeed, it has been suggested that ECMO is a “jack-of-all-trades” given its myriad applications (2).

However, experienced EMCO practitioners are intimately familiar with its limitations: high costs, intensive staffing, and potentially catastrophic complications. As a modality by which to accomplish end-organ perfusion, ECMO alone is not curative. The application of VA ECMO to cardiogenic shock can be conceptualized as a bridge to durable mechanical circulatory support, transplantation, recovery, or a decision when candidacy for definitive therapies is not immediately clear (3). For patients of advanced age, durable mechanical circulatory support and transplantation are often not realistic avenues (4). Indeed, such patients’ only ECMO escape strategy may be recovery. When recovery proves to be impossible or exceptionally unlikely, patients may then find themselves on a bridge to nowhere with associated distress for families and clinicians alike when faced with difficult decisions about stopping life-extending care that is intrinsically temporizing and not definitive (5,6). Given variability in ECMO practice patterns globally, it is difficult to conceptualize fully how and when ECMO is being used to support older adults and the associated outcomes.

The Extracorporeal Life Support Organization (ELSO) registry remains an important source of information to answer such questions. Through 2019, the registry contained data from 133,371 neonatal, pediatric, and adult ECMO runs as reported by 463 centers in 60 countries (7). When examining registry data, a larger number of total cases reported (e.g., as more centers participate) can confound the interpretation of trends in absolute counts. However, a previous analysis of ELSO registry data demonstrated that the proportion of patients 70 years old or older among all reported VA ECMO runs steadily increased over time from approximately 7% in the 1990s to 14.8% between 2011 and 2015 (8). Elderly patients had shorter ECMO runs and a 30.5% rate of survival to hospital discharge compared to 43.1% survival to discharge in the remainder of the adult cohort. A previous examination of the National Inpatient Sample to determine outcomes following VA ECMO for cardiogenic shock similarly demonstrated 32.3% survival to discharge in those 70–79 years old and 25.3% for those 80–90 years old (9).

In this issue of Critical Care Medicine, Kowalewski et al (10) build on that work and present an analysis of ELSO registry data from 1997 to 2018 to explore the outcomes of patients 70 years old or older who received VA ECMO support for refractory cardiogenic shock excluding extracorporeal cardiopulmonary resuscitation. Within the registry, 2,644 patients were identified, and the number of patients 70 years old or older within the registry increased markedly over time. As mentioned previously, this is somewhat difficult to interpret accurately without a relevant denominator of total registry cases but remains a striking finding regardless. Of this heterogeneous cohort, ECMO support was weaned successfully in 46.7% of patients with a 31.7% crude survival to discharge. In examining age as a categorical variable within a multivariable analysis, more advanced age (i.e., ≥ 80 yr vs 70–74 yr) was associated with higher mortality. The crude mortality of patients greater than or equal to 80 years old was 67.8%, which the authors noted to be roughly comparable with the 70–74 year-old (67%) and 75–79 year-old (70.1%) groups.

Of note, 58.9% of patients underwent VA ECMO for non-postcardiotomy shock (with over half of those having experienced acute coronary syndrome) while 41.1% had postcardiotomy shock (of which 41.3% could not be weaned from bypass). Given the broad indications for VA ECMO, this is a potentially important distinction as postcardiotomy shock is a different phenomenon affecting a different patient population than cardiogenic shock from other causes. For example, one might reasonably suspect patients with postcardiotomy shock to have less and/or better-controlled comorbid disease as they were selected to undergo surgery. Interestingly, there were no observed differences in mortality between the postcardiotomy and non-postcardiotomy cohorts. This may reflect the antecedent underlying influence of clinical decisions surrounding patient selection for VA ECMO (i.e., selection bias) inherent to the registry and the reality that practiced ECMO centers are the major contributors of data to the ELSO registry.

A few specific findings that may help to inform related clinical decision-making warrant emphasis. Of the examined pre-ECMO variables, need for renal replacement therapy was strongly associated with mortality. An interaction between age and the duration of support was observed, which is fully elucidated in the supplemental materials in (10). VA ECMO support beyond 192 hours, particularly in patients 75 years old or older, was associated with up to 90% mortality. Perhaps unsurprisingly, support for less than 96 hours was associated with both better survival and higher odds of stopping support on the basis of futility. This likely reflects the ability of VA ECMO to bridge successfully patients with cardiogenic shock to recovery who are impacted by pathophysiology that is both suspected to improve rapidly and then does actually improve. Conversely, patients of advanced age without near-term physiologic improvement may not go on to recover. Conceptually, eloquent editorialists have termed this a “short bridge over a wide river” (11). Additionally, clinicians may be using time-limited trials of ECMO support in patients of advanced age. When examining indications for support, heart transplantation, recent transcatheter aortic valve replacement, and pulmonary embolism were associated with improved survival, whereas sepsis and recent bypass plus valve replacement were not.

Kowalewski et al (10) have used their findings within the context of prior work to offer concise summary recommendations when considering support of elderly patients in refractory cardiogenic shock with VA ECMO, which is a challenging clinical dilemma. As decisions surrounding the initiation of ECMO are both difficult and often time-pressured, the importance of multidisciplinary input and well-defined internal criteria informed by evidence-based best practice has been previously recognized (12,13). The current work helps to further those goals.

The authors’ findings suggest that a carefully selected subset of adults over age 70 with refractory cardiogenic shock may have VA ECMO survival to discharge rates on-par with those of younger patients (i.e., approximately 45%); however, patient selection incorporating multidimensional data points remains paramount. Somewhat less restrictive selection criteria may conversely be associated with acceptable but comparatively worse survival to discharge (i.e., approximately 30%). These sobering figures offer helpful insight for clinicians when calibrating the expectations of families, care teams, and ourselves. In addition, these findings support the common clinical gestalt that lack of short-term physiologic improvement in older adults on ECMO may be a harbinger of unlikely eventual recovery.

Rather than viewing the present work as a call to expand the indications for ECMO in the most perilous of circumstances, it should further prompt us to examine our latent biases surrounding the role of age in clinical decision-making and reexamine the importance of patient-centered outcomes. Clinicians have demonstrated wide variability in their associated behaviors, and the critical examination of age-associated outcomes is one important step to combating this potential bias (14). Although the outcomes of refractory cardiogenic shock requiring VA ECMO may at best approximate 45% survival, the current work supports the growing notion that age alone should not be considered an absolute contraindication to extracorporeal support.

As the discipline of critical care medicine continues to expand its clinical and investigatory horizons beyond short-term outcomes, we have increasingly recognized that patients may suffer long-term deleterious sequelae of critical illness. Our conception of intermediate and long-term patient-centered outcomes following ECMO support is likewise in its nascency. The current work examines only inhospital mortality owing to the nature of the underlying registry dataset, which may itself speak to our priorities thus far. Given the material and existential risks of ECMO, better understanding its impact on intermediate and long-term patient-centered outcomes, including quality-of-life measures, remains an important avenue of investigation. This may be especially true when considering its application to those of advanced age who are at risk of impaired functional recovery after critical illness, namely, in the presence of frailty (15,16).

The work of Kowalewski et al (10) also challenges us to build systems of care through which critically important ECMO initiation decisions can be both well-informed but timely to seek continual improvement in short-term outcomes, to better support families and ourselves through clinically and morally difficult times, to acknowledge the risks of latent ageism underlying clinical judgment, and to better understand patient-centered intermediate and long-term functional outcomes.


1. Sanaiha Y, Bailey K, Downey P, et al. Trends in mortality and resource utilization for extracorporeal membrane oxygenation in the United States: 2008-2014. Surgery. 2019; 165:381–388
2. Kwak J, Majewski MB, Jellish WS. Extracorporeal membrane oxygenation: The new jack-of-all-trades? J Cardiothorac Vasc Anesth. 2020; 34:192–207
3. Guglin M, Zucker MJ, Bazan VM, et al. Venoarterial ECMO for adults: JACC scientific expert panel. J Am Coll Cardiol. 2019; 73:698–716
4. Chouairi F, Vallabhajosyula S, Mullan C, et al. Transition to advanced therapies in elderly patients supported by extracorporeal membrane oxygenation therapy. J Card Fail. 2020
5. Abrams DC, Prager K, Blinderman CD, et al. Ethical dilemmas encountered with the use of extracorporeal membrane oxygenation in adults. Chest. 2014; 145:876–882
6. Mulaikal TA, Nakagawa S, Prager KM. Extracorporeal membrane oxygenation bridge to no recovery. Circulation. 2019; 139:428–430
7. Extracorporeal Life Support Organization: ECLS Registry Report: International Summary. 2020. Available at: Accessed September 22, 2020
8. Lorusso R, Gelsomino S, Parise O, et al. Venoarterial extracorporeal membrane oxygenation for refractory cardiogenic shock in elderly patients: Trends in application and outcome from the Extracorporeal Life Support Organization (ELSO) registry. Ann Thorac Surg. 2017; 104:62–69
9. Chung M, Zhao Y, Strom JB, et al. Extracorporeal membrane oxygenation use in cardiogenic shock: Impact of age on in-hospital mortality, length of stay, and costs. Crit Care Med. 2019; 47:e214–e221
10. Kowalewski M, Zieliński K, Maria Raffa, et al. Mortality Predictors in Elderly Patients With Cardiogenic Shock on Venoarterial Extracorporeal Life Support. Analysis From the Extracorporeal Life Support Organization Registry. Crit Care Med 2020; 49: 7–18
11. Wolfe JD, Schilling JD. A short bridge over a wide river: The role of ECMO in older adults with cardiogenic shock. J Card Fail. 2020 Sep 9. [online ahead of print]
12. Abrams D, Garan AR, Abdelbary A, et al. International ECMO Network (ECMONet) and The Extracorporeal Life Support Organization (ELSO): Position paper for the organization of ECMO programs for cardiac failure in adults. Intensive Care Med. 2018; 44:717–729
13. DellaVolpe J, Barbaro RP, Cannon JW, et al. Joint society of critical care medicine-extracorporeal life support organization task force position paper on the role of the intensivist in the initiation and management of extracorporeal membrane oxygenation. Crit Care Med. 2020; 48:838–846
14. Garrouste-Orgeas M, Tabah A, Vesin A, et al. The ETHICA study (part II): Simulation study of determinants and variability of ICU physician decisions in patients aged 80 or over. Intensive Care Med. 2013; 39:1574–1583
15. Philippart F, Vesin A, Bruel C, et al. The ETHICA study (part I): Elderly’s thoughts about intensive care unit admission for life-sustaining treatments. Intensive Care Med. 2013; 39:1565–1573
16. Heyland DK, Garland A, Bagshaw SM, et al. Recovery after critical illness in patients aged 80 years or older: A multi-center prospective observational cohort study. Intensive Care Med. 2015; 41:1911–1920

cardiogenic shock; critical care outcomes; critical care; extracorporeal membrane oxygenation

Copyright © 2020 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.