Home About us Editorial board Ahead of print Current issue Search Archives Submit article Instructions Subscribe Contacts Login 
  • Users Online: 572
  • Home
  • Print this page
  • Email this page

 Table of Contents  
Year : 2016  |  Volume : 2  |  Issue : 1  |  Page : 22-26

Neurologic complications in patients receiving extracorporeal membrane oxygenation for influenza H1N1: Morbid but not futile

1 Department of Surgery, Summa Health System, Summa Akron City Hospital, Akron, Ohio, USA
2 Department of Internal Medicine, Summa Health System, Summa Akron City Hospital, Akron, Ohio, USA

Date of Submission26-Nov-2015
Date of Acceptance29-Jan-2016
Date of Web Publication2-Jun-2016

Correspondence Address:
Michael S Firstenberg
Department of Surgery (Cardiothoracic), Summa Akron City Hospital, 75 Arch Street, Suite 407, Akron, Ohio 44309, USA. Northeast Ohio Medical University, Rootstown, OH
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/2455-5568.183325

Rights and Permissions

Objectives: Extracorporeal membrane oxygenation (ECMO) is a well-established therapy for severe acute cardiac and respiratory distress. Management of complications, such as embolic strokes and intracranial hemorrhage, is essential during such treatment. However, the incidences of these complications as well as clinical outcomes have not been previously examined within a population with a homogenous disease process.
Methods: Using de-identified clinical data submitted to the Extracorporeal Life Support Organization, we analyzed cases in which patients were treated with ECMO during the 2009 H1N1 pandemic. This data were examined with a specific focus on neurologic complications.
Results: Twenty-two of the 248 patients experienced a confirmed neurological event. Patients with neurologic events were older, more acidotic, and had a higher prior incidence of cardiopulmonary arrest before ECMO therapy. There was also showed an increased incidence of cardiac arrhythmia, hyperbilirubinemia, and severe leukopenia as well as lower rates of successful weaning from ECMO and survival to discharge. Overall, within the patient population treated with ECMO, neurologic complications are not uncommon, and such patients exhibit greater morbidity and mortality.
Conclusions: Thus, aggressive neurological assessment before and during ECMO could prove very useful in guiding clinical decisions with respect to further or ongoing therapies. The findings from this study will hopefully improve patient selection for ECMO therapy as well as clinical outcomes for this critically ill patient population.
The following core competencies are addressed in this article: Patient care, Systems-based practice, Interpersonal skills and communication.

Keywords: Extracorporeal membrane oxygenation, influenza H1N1, respiratory failure, stroke

How to cite this article:
Hill S, Hejal R, Bowling SM, Firstenberg MS. Neurologic complications in patients receiving extracorporeal membrane oxygenation for influenza H1N1: Morbid but not futile. Int J Acad Med 2016;2:22-6

How to cite this URL:
Hill S, Hejal R, Bowling SM, Firstenberg MS. Neurologic complications in patients receiving extracorporeal membrane oxygenation for influenza H1N1: Morbid but not futile. Int J Acad Med [serial online] 2016 [cited 2023 Jan 31];2:22-6. Available from: https://www.ijam-web.org/text.asp?2016/2/1/22/183325

  Introduction Top

Extracorporeal membrane oxygenation (ECMO) is an established therapy for severe acute cardiac and/or respiratory failure. Recent data suggest improved outcomes in patients treated with ECMO for respiratory failure when compared to conventional maximal ventilator therapies.[1] As with any invasive therapy, patient selection and management of complications are critical to survival. Historically, neurologic complications, specifically embolic strokes, and intracranial hemorrhage, have been associated with poor outcomes in patients treated with ECMO for both cardiac and pulmonary failure.[2] However, recent data in patients undergoing ECMO support for respiratory failure with a homogenous disease process is poorly defined. Utilizing data voluntarily reported to an International Registry; we sought to characterize the incidence, risks, and short-term outcomes of patients sustaining neurologic complications in a contemporary series of patients with seasonal influenza H1N1. The specific aims of this study are to explore the incidence and outcomes of patients who are reported to have sustained a significant neurologic event while on ECMO. We hypothesize that such events are relatively uncommon and, when they do occur, are not inherently fatal.

  Methods Top

The Extracorporeal Life Support Organization (ELSO) is a voluntary international organization that maintains a registry of patients supported with ECMO for all causes.[3] Participating centers voluntarily submit data – including basic demographics, indications for support, duration of support, and self-reported complications, and outcomes of patients supported with ECMO at their Institution. The master registry tracks all self-reported adult, pediatric, and neonatal cases of ECMO (regardless of indications). It is the responsibility (and commitment) of each program to include accurate data on all patients supported with ECMO. Each program is provided summary and program-specific data for quality control and to assist in outcome reporting accuracy and program development. De-identified clinical data are submitted by participating centers to the organization in a manner compliant with local Institutional Review Board and data release practices. The University of Michigan Institutional Review Board provides registry oversight. Recognizing the growing use of ECMO in patients infected with H1N1 during the influenza pandemic of 2009, a separate registry was developed by ELSO to specifically track this homogenous population. Patients' data were voluntarily contributed by member centers to the H1N1-ECMO registry during 2009, 2010, and 2011. A total of 248 adults (>18 year/old) patients specifically treated with ECMO for confirmed influenza H1N1 from 160 international centers had data submitted. This dataset was retrospectively reviewed with specific focus on neurologic complications – specifically hemorrhagic bleeds or new embolic strokes as confirmed with CT scans of the head. Neurologic complications were self-reported by each site using ELSO predefined definitions as either binary events (new seizures: Yes/no) or descriptive (new stroke confirmed with computed tomography [CT]: Embolic, hemorrhagic, both). Other complications or events were also tracked using ELSO predefined definitions as either binary events (i.e. new infection, new dialysis on ECMO, death) or continuous variables (when appropriate).[3] As this data were self-reported and de-identified, there was no ability to blindly re-review or over-read imaging studies. All parameters examined are included in [Table 1]. It was the responsibility of each site to provide as accurate and complete data as possible.
Table 1: Patient characteristics

Click here to view

Student's t- and Fisher's tests were used to compare complication rates between patients with (S/B) and without (no S/B) neurologic embolic and/or hemorrhagic complications. A P < 0.05 was considered statistically significant. All data were provided by ELSO in a comma-delimited spreadsheet that was converted to a Microsoft Excel file (XLS Format, Microsoft Excel: 2008 for Mac, Microsoft Corporation, Redmond WA, USA). Descriptive data were summarized using built-in spreadsheet tools (i.e. averages, standard deviations, minimums, maximum). Statistical testing (i.e. Fisher's 2 × 2 contingency testing) was performed using Internet-based, on-line tools (http://graphpad.com/quickcalcs/contingency1.cfm).

  Results Top

Of the 248 patients included in the registry, five patients sustained new, CT scanning confirmed, embolic strokes, 15 had confirmed hemorrhagic events, and 2 had both. Overall, 22 patients (8.9%) sustained a documented neurologic event. [Table 1] outlines the clinical characteristics of those patients sustaining a significant neurologic event while on ECMO when compared to those who did not. In general, patients who sustained a neurologic event were significantly older and more acidotic before ECMO therapy (both P < 0.05) and tended to have been more likely to have sustained a cardiopulmonary arrest before initiation of ECMO (P = 0.08). While on ECMO, neurologic events were more likely in patients who developed cardiac arrhythmias, hyperbilirubinemia, and severe leukopenia (white blood cell count <1.5k cells/µL) (all P < 0.05). In addittion, significant was the lower rate of successful weaning from ECMO and being discharged alive in patients with neurologic events.

However, there was no significant difference in body weight between the two groups. There was no difference in genders, despite 68% of the S/B group being male with only 51% in the no S/B group (P = 0.18). Importantly, the time between initial intubation and initiation of ECMO therapy was not significant between the S/B and no S/B groups, nor was the duration of ECMO support within each group. There was not a significant difference in pre-ECMO mean blood pressures between the S/B and no S/B groups.

In addition, S/B's were not associated with ECMO mechanical circuit or pump malfunctions, the development of a new infection (of any type), or renal complications (including the need for renal replacement therapy) [Table 1], all P > 0.05].

  Discussion Top

Seasonal Influenza H1N1 and more recent genetic strains, such as H5N1, have been associated with acute respiratory distress syndrome (ARDS).[4] The pathologic impact of H1N1 during 2009 was significant as this strain was suspected or confirmed in as many as 25% of all patients in the intensive care units in New Zealand and Australia during the peak of the pandemic.[5] However, unlikely previous seasonal flu epidemics that primarily impacted vulnerable patients at the extreme of age, or with multiple severe co-morbidities, H1N1 appeared to have a disproportionate case fatality rate on otherwise healthy, relatively young, and adults.[6],[7] Because many of these patients were otherwise young and healthy, clinicians aggressively utilized ECMO as salvage therapy.[8]

Historically, ECMO was often used in salvage cases when all other treatment options have been exhausted, and death would otherwise ensue – and, unfortunately, outcomes were poor.[9],[10] However, recent evidence and changes in therapy paradigms suggest that when used early before irreversible end-organ damage and refractory shock, ECMO can result in more favorable outcomes.[11] Despite the increasing use of ECMO, its neurological effects have not been well studied, and publications on this topic are limited.[10] Previous studies, typically retrospective and single-center series, have demonstrated neurologic complications in patients treated with ECMO have often included a variety of indications or therapy and provide little insight into the patient characteristics, incidence, and outcomes of such events. While the need for systemic anticoagulation and the concern of pre-ECMO hypoxic-ischemic neurologic injuries are often viewed as key contributing factors, recent experiences suggest other, potentially modifiable, variables as causative agents of embolic or hemorrhagic events. Lower nadir arterial oxygen pressure while on ECMO, acidosis (metabolic or respiratory), renal failure, the utilization of cardiopulmonary resuscitation while on ECMO, the method of ECMO support used (veno-veno vs. veno-arterial) all convey an increased risk for poor overall outcomes.[9],[10] In addition, recent experimental data suggests that the rapid normalization of arterial CO2 tension (PaCO2) accomplished by initiation of ECMO can result in focal areas of impaired cerebral oxygenation and a predisposition to intracranial bleeding.[12] A significant advantage of this study is that it reviews the experiences and outcomes with a homogenous disease condition (influenza H1N1) with a predicted base-line neurologic complication rate that is similar to what we experienced in patients who required ECMO. Previous studies have suggested 4% neurologic complication rate of H1N1, but most of these are seizures and encephalopathies presumed to be related to associated hypoxemia. Strokes and bleeds in this experience with 2069 severe or fatal H1N1 cases (unclear how many, if any, were treated with ECMO) not reported.[13] These findings suggest that the stroke and bleed rates, while on ECMO, might be more likely related to ECMO therapy than the disease process being treated.

Given the difficulty related to monitoring neurological function during ECMO treatment, the true incidence of brain injury may never be completely known. Patients are often heavily sedated and pharmacologically paralyzed, difficult to transport for imaging, and/or die of other complications in the absence of a complete neurologic assessment. In addition, as patients are often extremely ill before initiation of emergent ECMO therapy, a comprehensive neurologic evaluation is often not performed. Nevertheless, techniques are available to assist in the bedside monitoring of patients and to help guide decision-making. It has been proposed that tools routinely used in the intensive care setting (i.e. non-ECMO patients) can be applied to patients being considered for ECMO and who are undergoing treatment.[14] Bedside testing, including electroencephalograms with somatosensory evoked potentials to monitor cerebral blood flow, transcranial Doppler, jugular oxygen saturation, and near-infrared spectroscopy can be useful in detecting pathologic brain function from all causes. However, clinical utility in patients treated with ECMO is poorly understood. Nevertheless, alterations in these test results could rapidly and significantly impact patient management – including obtaining CT scanning, initiating antineuroleptic therapies, or even pursuing conversations with families regarding the long-term goals of therapy. Routine monitoring might even help better define the scope neurologic complications in this patient population. Without a doubt, a pupillary examination and a comprehensive neurological physical examination at the bedside in a comatose patient should be completed whenever possible before paralysis and deep sedation, before or during ECMO.

Specifically related to our findings, a greater incidence of acidosis was observed in the S/B group. However, this may be seen as an indication that these patients were in greater need of oxygenation, and thus, the need for ECMO intervention was more critical, but not necessarily a contraindication for utilizing ECMO. A low pH indicates lower oxygen and higher carbon dioxide content in the blood. Systemic hypoxemia could predispose an individual to neurological events due to ischemia as the brain requires a constant supply of oxygen to maintain tissue integrity as well as function.[15] A greater incidence of pre-ECMO arrest was also observed in the S/B group compared to the no S/B group. While this difference only trends toward statistical significance (P = 0.08), it again supports the theory that acidosis or any degree of impaired cerebral perfusion might precipitate a neurologic event in patients treated with ECMO. The S/B patients exhibited a greater number of complications such as arrhythmias (P < 0.001), hyperbilirubinemia (P = 0.02), and white blood cell count <1.5k cells/µL (0.04), compared to the no S/B patients. These findings suggest a greater degree of end-organ dysfunction and whether they are markers for potential neurologic events or causative variable is unclear – nevertheless, our findings suggest that such critically ill patients might benefit from more intensive neurologic evaluations before and after initiating ECMO.

Perhaps most concerning, although not surprising, is the stark contrast between the percentage of S/B patients and no S/B patients weaned off ECMO, as well as overall survival rate. Far fewer of the S/B patients were weaned off ECMO (41%), with only 18% total discharged alive. Even regarding those who were weaned off of ECMO, less than 44% of the patients who experienced neurological complication survived to discharge. The no S/B group had a significantly improved outcome, in comparison, with a majority weaned off of ECMO (79%), with 71% total discharged alive. This data show the dramatic impact of neurologic events on the ability to be successfully weaned from ECMO and subsequently discharged alive. Clearly, cause and effect relationship is unclear as patients sustaining neurologic events while on ECMO might be more likely to have support withdrawn before attempts at weaning or, conversely, might have ECMO discontinued before the adequate recovery of lung function and hence, predispose the patient to death from pulmonary failure. However, despite these associations, our findings clearly demonstrate that the overall documented incidence of embolic strokes or cerebral hemorrhages in patients on ECMO is relatively low. As importantly, when they do occur, while the outcomes tend to be worse, such complications are not inherently fatal or suggest further therapies as being futile.


Our findings have numerous limitations that need to be considered when evaluating our findings. Clearly, the data available was not only retrospective but also voluntary, with little ability to validate the overall findings. While there is no inherent reason to question to integrity or completeness of the data, as discussed above, the limitations of obtaining detailed neurologic assessments and imaging on patients before or during treatment with ECMO cannot be ignored. The risk of recall bias and the reporting of only positive outcomes and under-reporting complications clearly exists. Furthermore, as this was H1N1 registry data, numerous data fields that were not collected that might, in retrospect, provide a greater insight into this clinical problem. No data regarding patient co-morbidities, including pre-ECMO neurologic status, was available to help determined confounding predictive variables. In addition, since indications for initiating therapy, cannulation, and anticoagulation techniques, and even pump management might varying considerable across the reporting centers extrapolation of our finding to routine clinical practice must be taken with caution particularly when considering whether to deny or withdraw life-saving therapies.

Despite these challenges, these findings probably reflect the largest single disease-state (i.e. Influenza H1N1) dataset exploring the question of neurologic complications in patients undergoing ECMO therapy for respiratory failure. Our findings and the limitations of this study, as with any retrospective registry review, illustrate the need to further explore, in a comprehensive and prospective fashion, those variables associated with neurologic complications.

  Conclusions Top

In this homogenous H1N1 patient population, treated with ECMO for refractory respiratory failure, neurologic complications were not uncommon. This study confirms that patients with neurologic complications have greater morbidity and mortality while on ECMO. Aggressive neurological assessment – both before and while on ECMO support – may be helpful in guiding clinical decisions regarding further or on-going therapies. Although limited by patients who may not have been diagnosed, S/B's were associated with an increased risk of inability to wean from ECMO and subsequent death. However, not all patients experiencing a S/B died. These findings can hopefully further optimize patient selection, family communication, and improve outcomes in this high-risk, resource intensive, critically ill patient population.

This work was supported by a research grant provided by the Extracorporeal Life Support Organization, and Ms. Hill was supported as Summa Health Cardiothoracic Surgical Summer Scholar from an educational grant from the Department of Surgery and Summa Research Foundation.

Financial support and sponsorship

This work was supported by a research grant provided by the Extracorporeal Life Support Organization, and Ms. Hill was supported as Summa Health Cardiothoracic Surgical Summer Scholar from an educational grant from the Department of Surgery and Summa Research Foundation.

Conflicts of interest

There are no conflicts of interest.

  References Top

Peek GJ, Mugford M, Tiruvoipati R, Wilson A, Allen E, Thalanany MM, et al. Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): A multicentre randomised controlled trial. Lancet 2009;374:1351-63.  Back to cited text no. 1
Brodie D, Bacchetta M. Extracorporeal membrane oxygenation for ARDS in adults. N Engl J Med 2011;365:1905-14.  Back to cited text no. 2
ELSO Website Homepage. Available from: . [Last accessed on 2016 Jan 20].  Back to cited text no. 3
Luyt CÉ, Combes A, Trouillet JL, Nieszkowska A, Chastre J. Virus-induced acute respiratory distress syndrome: Epidemiology, management and outcome. Presse Med 2011;40 (12 Pt 2):e561-8.  Back to cited text no. 4
Australia and New Zealand Extracorporeal Membrane Oxygenation (ANZ ECMO) Influenza Investigators, Davies A, Jones D, Bailey M, Beca J, Bellomo R, et al. Extracorporeal membrane oxygenation for 2009 influenza A (H1N1) acute respiratory distress syndrome. JAMA 2009;302:1888-95.  Back to cited text no. 5
Domínguez-Cherit G, Lapinsky SE, Macias AE, Pinto R, Espinosa-Perez L, de la Torre A, et al. Critically ill patients with 2009 influenza A (H1N1) in Mexico. JAMA 2009;302:1880-7.  Back to cited text no. 6
Moreno RP, Rhodes A, Chiche JD. The ongoing H1N1 flu pandemic and the intensive care community: Challenges, opportunities, and the duties of scientific societies and intensivists. Intensive Care Med 2009;35:2005-8.  Back to cited text no. 7
Azevedo LC, Park M, Costa EL, Santos EV, Hirota A, Taniguchi LU, et al. Extracorporeal membrane oxygenation in severe hypoxemia: Time for reappraisal? J Bras Pneumol 2012;38:7-12.  Back to cited text no. 8
Mehta A, Ibsen LM. Neurologic complications and neurodevelopmental outcome with extracorporeal life support. World J Crit Care Med 2013;2:40-7.  Back to cited text no. 9
Mateen FJ, Muralidharan R, Shinohara RT, Parisi JE, Schears GJ, Wijdicks EF. Neurological injury in adults treated with extracorporeal membrane oxygenation. Arch Neurol 2011;68:1543-9.  Back to cited text no. 10
Richard C, Argaud L, Blet A, Boulain T, Contentin L, Dechartres A, et al. Extracorporeal life support for patients with acute respiratory distress syndrome: Report of a consensus conference. Ann Intensive Care 2014;4:15.  Back to cited text no. 11
Kredel M, Lubnow M, Westermaier T, Müller T, Philipp A, Lotz C, et al. Cerebral tissue oxygenation during the initiation of venovenous ECMO. ASAIO J 2014;60:694-700.  Back to cited text no. 12
Glaser CA, Winter K, DuBray K, Harriman K, Uyeki TM, Sejvar J, et al. A population-based study of neurologic manifestations of severe influenza A (H1N1) pdm09 in California. Clin Infect Dis 2012;55:514-20.  Back to cited text no. 13
Zanatta P, Bosco E, Forti A, Polesel E, Sorbara C. Neurological monitoring during ECMO. In: ECMO-Extracorporeal Life Support in Adults, F Sangalli, N Patroniti(, Pesenti A. (eds). Springer-Verlag Italia, Milan, I.T. 2014, pp. 389-99.  Back to cited text no. 14
Zauner A, Daugherty WP, Bullock MR, Warner DS. Brain oxygenation and energy metabolism: Part I-biological function and pathophysiology. Neurosurgery 2002;51:289-301.  Back to cited text no. 15


  [Table 1]


Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

  In this article
Article Tables

 Article Access Statistics
    PDF Downloaded93    
    Comments [Add]    

Recommend this journal