Skip Navigation
Skip to contents

ACC : Acute and Critical Care



Page Path
HOME > Acute Crit Care > Volume 38(4); 2023 > Article
Review Article
Plasma biomarkers for brain injury in extracorporeal membrane oxygenation
Shrey Kapoor1orcid, Anna Kolchinski1orcid, Aaron M. Gusdon2orcid, Lavienraj Premraj3,4orcid, Sung-Min Cho1orcid
Acute and Critical Care 2023;38(4):389-398.
Published online: November 29, 2023

1Division of Neurosciences Critical Care and Cardiac Surgery, Departments of Neurology, Surgery, Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA

2Division of Neurocritical Care, Department of Neurosurgery, McGovern School of Medicine, University of Texas Health Science Center, Houston, TX, USA

3Griffith University School of Medicine, Queensland, Australia

4Critical Care Research Group, The Prince Charles Hospital, Queensland, Australia

Corresponding author: Sung-Min Cho Division of Neurosciences Critical Care and Cardiac Surgery, Departments of Neurology, Surgery, Anesthesiology and Critical Care Medicine, Johns Hopkins University School of Medicine, 600 N. Wolfe St, Phipps 455, Baltimore, MD 21287, USA Tel: +1-513-582-3009 Email:
*These authors contributed equally to this work.
• Received: October 27, 2023   • Revised: November 24, 2023   • Accepted: November 24, 2023

Copyright © 2023 The Korean Society of Critical Care Medicine

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 83 Download
  • Extracorporeal membrane oxygenation (ECMO) is a life-saving intervention for patients with refractory cardiorespiratory failure. Despite its benefits, ECMO carries a significant risk of neurological complications, including acute brain injury (ABI). Although standardized neuromonitoring and neurological care have been shown to improve early detection of ABI, the inability to perform neuroimaging in a timely manner is a major limitation in the accurate diagnosis of neurological complications. Therefore, blood-based biomarkers capable of detecting ongoing brain injury at the bedside are of great clinical significance. This review aims to provide a concise review of the current literature on plasma biomarkers for ABI in patients on ECMO support.
Over the past 30 years, extracorporeal membrane oxygenation (ECMO) has been increasingly deployed to provide life-saving circulatory support to patients with refractory pulmonary and cardiac pathology [1]. Venoarterial (VA)-ECMO is used as the rescue of refractory cardiac failure and venovenous (VV)-ECMO is used for refractory respiratory failure. Despite its benefits, current literature indicates an overall ECMO survival rate of 29%, with acute brain injury (ABI), including acute ischemic stroke (AIS), intracranial hemorrhage (ICH), and hypoxic-ischemic brain injury, being a leading cause of morbidity and mortality [2]. Further, ABI confers a two- to three-fold increase in mortality risk for patients on ECMO [2-4]. Standardized neuromonitoring protocols improve the detection of ABI and include iterative neurological exams, near-infrared spectroscopy, transcranial doppler, electroencephalograms, and somatosensory evoked potential and neuroimaging studies (Figure 1) [3,4].
However, neuroimaging during ECMO is challenging. Head computed tomography (CT) scans lack sensitivity to detect early and small ischemic events. Moreover, head CT scans are obtained in only 30% of ECMO patients due to the increased risks of transporting patients while cannulated for ECMO [3]. Brain magnetic resonance imaging (MRI) is more sensitive for identifying ischemia but MRI with high-strength magnetic fields (1.5–3T) is incompatible with ECMO [5]. Additionally, most patients receiving ECMO support would not be stable enough to undergo a prolonged imaging study outside of the intensive care unit. To overcome these limitations, the development and validation of accessible plasma biomarkers that are sensitive and specific for ABI are necessary. Consequently, there has been a recent interest in plasma biomarkers for patients on ECMO (Figure 2). The purpose of this review is to concisely consolidate and critically appraise our current understanding of plasma biomarkers and their role in monitoring and prognostication of ABI and their applicability in patients on ECMO support.
Biomarkers in VA-ECMO

Neuron-specific enolase

Neuron-specific enolase (NSE) is a glycolytic enzyme prevalent in neurons and neuroendocrine cells [5]. NSE bloodstream levels, which begin to increase several hours after neuronal damage and peak at 48 hours post-injury, are a potential biomarker for brain injury in patients undergoing ECMO [6,7]. Several studies have focused on elucidating the role of NSE in various clinical settings. A prospective multicenter study involving 623 out-of-hospital cardiac arrest (OHCA) survivors revealed a significant difference in median NSE levels between patients with non-shockable and shockable rhythms (104.6 ng/ml [range, 40.6–228.4] vs. 25.9 ng/ml [range, 16.7–53.4], P<0.001) [6]. In a retrospective multicenter study on 70 traumatic brain injury (TBI) patients, higher NSE concentrations corresponded with increased injury severity and adverse neurological outcomes [8].
The utility of NSE as a biomarker for patients on ECMO is still under investigation. A prospective study involving 80 pediatric ECMO patients indicated elevated NSE levels in those developing neurological complications [9]. Specifically, elevated plasma concentrations of NSE, with a cutoff of 62.0 ng/ml, were associated with increased odds (adjusted odds ratio, 2.89) of unfavorable outcomes, measured by Pediatric Cerebral Performance Category, at hospital discharge [9]. These findings are consistent with another retrospective single-center study of 30 adults in a VA-ECMO cohort, which showed that NSE peak values above a cutoff of >100 μg/L had a sensitivity and specificity for mortality of 0.53 and 0.82, respectively [10].
However, it’s important to highlight that the specificity of NSE in ECMO patients may be compromised by erythrocyte hemolysis, a frequent occurrence in ECMO, which can lead to elevated NSE levels, complicating its reliability in detecting ABI and predicting neurological outcomes [11]. Future prospective multicenter research is necessary to understand specific NSE thresholds and dynamic kinetics in relation to ABI in ECMO patients.

Glial fibrillary acidic protein

Glial fibrillary acidic protein (GFAP) is a specific type of intermediate filament protein, involved in the structural integrity and functioning of astrocytes within the central nervous system (CNS). The unique localization of GFAP to astrocytes in the brain emphasizes its potential as a sensitive biomarker for detecting brain injuries. Astrocytes undergo astrogliosis—a reactive process involving cellular proliferation and increased GFAP production—indicating neuronal damage [12]. GFAP is elevated within about 2 hours following injuries such as AIS [13-15], cardiac arrest [16,17], and TBI [18-22], peaking at 48–72 hours [23], and thus may be useful for monitoring both neurologic and non-neurologic complications in ECMO.
In a pediatric single center VA-ECMO cohort (n=80), GFAP levels were measured by immunoassay at 6 and 12 hours, and then every 24 hours after cannulation [9]. GFAP levels were found to be significantly elevated in children on ECMO who experienced ABI compared with children who did not, with a median of 5.9 versus 0.09 ng/ml (P=0.04) [9]. Additionally, higher GFAP levels were found in non-survivors compared to survivors. GFAP levels over 0.436 ng/ml were associated with increased odds ratio (OR) of brain injury (OR, 11.5; 95% confidence interval [CI], 1.3–98.3) and mortality (OR, 13.6; 95% CI, 1.7–108.5) [9]. However, this study was limited by the small sample size and the difficulty of accurately diagnosing ABI. A pilot single-center study by Kapoor et al. [24] (n=20) corroborated these findings in adults on VA-ECMO support, with median plasma GFAP levels noted to be significantly higher for patients who developed ABI vs. without (5.17 ng/ml vs. 0.16 ng/ml, P=0.047). The findings from these pilot studies warrant further research in larger cohorts to validate and define their clinical applications.


Tau proteins are microtubule-associated proteins involved in stabilizing the cytoskeleton, particularly in unmyelinated axons of the CNS [25]. While they are most abundant in neurons, they are also found in astrocytes and oligodendrocytes [25]. Like neurofilament light chain (NfL), tau protein released into the cerebrospinal fluid (CSF) can potentially be used as a biomarker indicating axonal damage. Tau has been most heavily studied in its misfolded aggregate form in Alzheimer disease and other forms of dementia, such as chronic traumatic encephalopathy in athletes playing contact sports [26,27].
Elevated tau has shown promise as a marker for adverse clinical outcomes following TBI. In a pilot, single-center prospective study of 16 patients, microdialysis catheter measurement of tau from the brain extracellular space following TBI correlated with a lower Glasgow Coma Scale (GCS) score 6 months after injury (r=–0.6, P=0.018) [28]. Additionally, tau levels were highest when catheters were near the area of the contusion, verifying the use of tau as a marker of axonal damage [26]. These conclusions were strengthened by another single-center prospective study (n=50), which showed that CSF tau levels correlated with AIS severity (P<0.05). In this study, higher CSF tau levels were also correlated with a worse outcome defined by the modified Rankin Scale (mRS) score at 12 months [29]. In an international multicenter prospective study (n=689), increased serum tau soon after OHCA was predictive of poor neurological function at 6 months, as measured by the Cerebral Performance Categories (CPC) scale. At 48 hours following cardiac arrest, the median tau levels were 49.5 ng/L in the poor outcome group, and only 1.2 ng/L in the favorable outcome group (P<0.0001) [30]. Tau levels may be useful for more extended neuromonitoring, as while elevations begin as early as 6 hours post-injury, peak elevations are as late as 6 days following injury [31].
Despite the promising results of tau protein predicting outcomes in neurological diseases, evidence is very limited in ECMO patients. A pilot single-center prospective observational cohort study by Kapoor et al. [24] (n=20) reported a higher trend in plasma tau values for patients with ABI vs without ABI on VA-ECMO (1.87 ng/ml vs. 0.44 ng/ml, P=0.62). This study was limited by its sample size and was unable to reach appropriate power for statistical significance. While tau shows promise as a biomarker after TBI and cardiac arrest, its utility for patients on ECMO needs to be further explored.

Neurofilament light chain

Neurofilaments are intermediate filaments that help comprise the cytoskeleton of neurons. The triplet consists of three subunits, including NfL, neurofilament medium chain, and neurofilament heavy chain [32]. NfL is a neuron-specific protein residing in the cytoplasm of large, myelinated axons [32]. NfL has been studied as a potential biomarker, as its release into the CSF as early as several hours after injury can indicate neuronal death and more specifically axonal damage [33,34]. To date, NfL has been most thoroughly studied in multiple sclerosis, TBI, and cardiac arrest, and its application in other neurologic disorders has recently been explored including stroke, lysosomal storage disorders, amyotrophic lateral sclerosis, and frontotemporal dementia.
NfL has shown promise as a marker for outcome prediction in patients with OHCA. An international, multi-center (29 sites) prospective study of 782 adults demonstrated that serum NfL levels measured 1–3 days after OHCA were sensitive and specific for worse neurological outcomes by CPC score at 6 months, and this prognostic performance exceeded that of other biomarkers and standardized neuromonitoring (CT, somatosensory evoked potentials, electroencephalogram, and neurological exam) [35]. A multicenter prospective study of 248 adults validated these findings and found that median serum NfL concentrations 48 hours after OHCA was 0.019 ng/ml in patients with good neurological outcomes by CPC score and 2.343 ng/ml in those with poor neurological outcomes (P<0.001) [36]. In a retrospective multicenter observational study (n=428) looking at CPC outcomes for both OHCA and in-hospital cardiac arrest (IHCA), NfL at 12 hours was superior to NSE at 24 hours (area under the curve [AUC]: 0.93 vs. 0.83 for OHCA and AUC: 0.87 vs 0.71 for IHCA) [37]. NfL at 48 hours was superior to NSE at 48 hours (AUC: 0.97 vs. 0.90 for OHCA, and AUC: 0.90 vs. 0.77 for IHCA) [37].
A pilot prospective single-center cohort study (n=20) reported a significant difference in NfL values for patients with ABI vs without ABI on VA-ECMO (13.6 ng/ml vs. 3.42 ng/ml, P=0.007), which may be better than GFAP and tau protein in predicting ABI and neurological outcomes in ECMO patients [24]. In summary, serum NfL appears to be the most promising biomarker with consistent demonstration of high sensitivity and specificity across various forms of brain injury. Further studies to determine accurate cutoffs are warranted for ECMO patients.
Biomarker Applications Studied in VV-ECMO


MicroRNAs (miRNAs) are a class of single-stranded, non-coding RNA fragments involved in regulating gene expression in many pathways [38]. While currently difficult to implement miRNA in a critical care setting due to the cost and time needed to determine miRNA sequencing signatures, a prospective study in 11 adult VV-ECMO patients with severe acute respiratory distress syndrome (ARDS) found that serum levels of 13 miRNAs in were markedly upregulated (vs. those with less severe disease) as measured by Sequential Organ Failure Assessment (SOFA), respiratory ECMO survival prediction score, and Simplified Acute Physiology Score (SAPS) II [38]. miRNAs increased in severe ARDS were found to be members of tissue remodeling, blood coagulation, and immune regulation pathways, in line with processes found to contribute to further injury in ARDS patients [38]. While not specifically studying neurological outcomes, as miRNA sequencing becomes more available and cost-effective, similar miRNA upregulation signatures could be identified specifically for neurological injury. The breadth of available miRNAs means that biomarkers could be unique according to tissue type, and even to processes such as ischemic and hemorrhagic stroke.
Biomarker Applications Studied in VA-ECMO and VV-ECMO

Inflammatory cytokines

Inflammatory cytokines such as interleukin (IL)-6, IL-1, IL-8, and tumor necrosis factor (TNF) -α are small, secreted molecules controlling cell-cell interaction in the setting of inflammation regulation [39]. While it has been well-established that inflammatory cytokines rise throughout the duration of ECMO therapy [40], few studies have focused on inflammatory cytokines as predictors of neurological injury in the setting of ECMO. One prospective pediatric study of 99 patients undergoing both VA-ECMO and VV-ECMO found significant elevations in inflammatory markers including interferon gamma (IFN-γ), IL-6, IL-1β, and TNF-α in children with abnormal head imaging (diffuse hypoxic injury, ICH, or ischemic stroke on ultrasound, CT or MRI) compared to children with normal head imaging. Additionally, IL-6 concentration was significantly elevated in children who did not survive their hospitalization compared to children who did, both on ECMO day 1 (median, 98.5 vs. 23.1 pg/ml) and at peak (median, 102.8 vs. 32.6 pg/ml) [41].
A prospective observational cohort study (n=266) on inflammatory cytokines in adults undergoing VV-ECMO found that higher inflammatory cytokines during ECMO were associated with a lower chance of survival, suggesting that increased inflammation during ECMO may be a potential prognostic biomarker [42]. This study also showed that levels of inflammatory cytokines including IL-6, IL-8, and TNF-α fell after discontinuing mechanical ventilation and switching to ECMO [42]. While mechanisms were not fully explored, patients undergoing mechanical ventilation with positive end-expiratory pressure (PEEP) ≥15 cm H2O and/or driving pressures ≥ 19 cm H2O prior to ECMO had higher inflammatory cytokine levels than those who had required lower PEEP and driving pressures [42]. Oncostatin M (OSM) is an inflammatory cytokine in the IL-6 family that is expressed in neurons, astrocytes, glia, and immune cells infiltrating the CNS, but also in non-neural tissue, such as lung and liver [43]. However, the role of OSM in ABI is controversial, with conflicting evidence. A prospective study of 29 VV-ECMO patients found that pre-decannulation OSM was elevated in VV-ECMO patients who did not survive compared to those who did (Mann-Whitney U-test: P=0.04) [43]. However, prior murine models have demonstrated that increased OSM expression is associated with improved neurological recovery following AIS, in contrast to the VV-ECMO study [44]. While studies of OSM after ABI have been contradictory, prior studies in lung injury have shown that OSM is released by infiltrating immune cells, with increased OSM serving as a marker of tissue injury [45]. However, given that OSM is not specific to neurological injury, and prior results have been conflicting, OSM may not be as useful compared to other brain-specific biomarkers that are discussed in this review.
Overall, inflammatory cytokines as an umbrella category are of variable utility as predictors of neurological injury in the setting of ECMO. Inflammatory cytokines such as IFN-γ, IL-6, IL-1β, and TNF-α have shown utility as an indicator of neurological complications in the pediatric ECMO population and as predictors of mortality in the adult population and require further study to establish reference ranges in the setting of various types of neurological injury. Conversely, OSM has shown conflicting results in various studies and lacks neurological specificity, and thus its utility in indicating neurological injury during ECMO may be limited.
Mitochondrial DNA
Mitochondrial DNA (mtDNA), sometimes referred to as circulating free mtDNA, fragments are released when cells are affected by mechanical or biological injury (as in hypoxia or inflammation) [46]. Physiologically, mtDNA appears to potentiate the immune response via toll-like receptors [46-48]. mtDNA levels have previously shown utility in predicting disease severity and mortality in critically ill patients with sepsis and acute lung injury [49,50].
In the setting of mechanical circulatory support, investigations into the role of mtDNA are limited. In a prospective study of 962 adult patients with post-cardiopulmonary bypass, mtDNA levels positively correlated with N-terminal prohormone of brain natriuretic peptide (NT-proBNP) levels and Acute Physiology and Chronic Health Evaluation (APACHE) II score, an estimate of intensive care unit mortality risk [51]. In an ex vivo ECMO model that circulated heparinized human blood for 6 hours at differing flow rates, mtDNA was associated with loss in platelet function as measured by thrombelastography platelet mapping and collagen agonist stimulated responses [52]. mtDNA levels may have utility in predicting coagulation complications in ECMO patients, particularly in patients with ischemic and hemorrhagic strokes. In a single-center prospective study (n=100), mtDNA levels were significantly higher in patients with AIS compared to controls (P<0.05), positively correlated with clinical severity of stroke as determined by the National Institute of Health Stroke Scale, and that plasma mtDNA levels decreased with time after stroke (follow-up for up to 30 days, P<0.001) [53].
In summary, mtDNA is a promising biomarker specifically for patients with coagulation complications in ECMO. Further research specifically in these cohorts across centers is necessary to validate these findings and determine accurate cutoffs.
Cell-Free DNA
Cell-free DNA (cfDNA) is a novel biomarker consisting of double-stranded DNA fragments released into the bloodstream under conditions of tissue stress resulting in cellular apoptosis or necrosis [54]. While cfDNA has been studied primarily in current literature in its application in liquid biopsy for oncological diagnosis and management, there has been a recent uptick in understanding the role that cfDNA may play in acute neurological conditions. In a prospective study of 54 patients with AIS, cfDNA was elevated in blood samples collected within 48 hours of AIS treated with IV thrombolysis compared to healthy controls [55]. Further, cfDNA levels on admission were able to discriminate between AIS and stroke mimics. An analysis of the plasma beta-globin gene allowed discrimination between AIS and ICH [55]. Similarly, in a prospective study of 54 Indian patients with AIS, cfDNA level was shown to correlate with poor neurological outcomes (measured by mRS score) at 3 months [56]. Additionally, in a prospective study of 60 patients admitted after acute ICH, higher plasma nuclear DNA levels at presentation correlated with poorer GCS scores [57]. Plasma cfDNA levels have also been shown to be predictive not only of worse functional outcomes, but also of stroke size at onset. In a prospective study of 88 patients presenting to the emergency department with stroke-like symptoms, plasma cfDNA taken within 3 hours of symptom onset correlated with ICH volume as measured by CT [58]. Although there are no studies using cfDNA in ECMO patients, given the utility of cfDNA in predicting stroke volume and severity, plasma cfDNA levels may be useful as an early indicator of possible stroke in ECMO patients.
Lipidomics and Metabolomics
Lipidomics and metabolomics offer a promising new avenue for markers of TBI, due to the increased permeability of the blood brain barrier (BBB) to lipids compared to proteins [59]. Cellular injury occurring after TBI carries with it a plethora of metabolic effects, resulting in derangements in processes such as oxidative phosphorylation and activation of phospholipase C and D, thereby triggering downstream signaling cascades. When the rat serum lipidome was analyzed 3 and 7 days after a controlled cortical TBI, lipidome panels achieved 85% accuracy at differentiating rats having undergone TBI from the control rats. Amongst lipids analyzed, polyunsaturated fatty acids (PUFAs), specifically free fatty acids, increased after TBI, while oxidized fatty acids decreased. Additionally, PUFA-esterified diacylglycerols increased, due to phospholipase signaling [59]. In further work by the same group, sensitivity of the lipidome panel exceeded 90% for lipidomic changes occurring within 24 hours following TBI. Sphingolipids, specifically sphingomyelins (SM), were increased as early as 4 hours following the injury, due to increased cell death. Additionally, ceramides were found to be increased, due to increased necroptosis [60].
In a metabolomic cohort study of 716 patients with TBI, carbohydrates derivatives including myoinositol, found in glial cells, were found to be elevated following TBI, which was postulated to be due to disruption of the BBB. Conversely, higher levels of phospholipids, including phosphatidylcholines, lysophosphatidylcholines, and SM, were correlated with decreased severity of TBI as well as improved functional outcomes [61]. Higher circulating levels of lysophospholipids were also found to be associated with improved outcomes in patients with mid TBI [62]. These results highlight the complex effects of circulating phospholipids and significant differences in the systemic response to ABI in humans compared with rodent models. Further research is needed to better characterize the functional consequences of altered lipid levels after ABI before they can be used effectively as clinical biomarkers.
While lipidomic and metabolomic changes have not yet been correlated to neurological complications of ECMO, the stability of lipids and their easier passage across the BBB make them promising candidates as biomarkers of ABI. Given the few numbers of studies, more work is needed in human subjects to characterize both lipidome and metabolome derangements, as well as their pathophysiology, prior to application in patients at risk for ABI including those on ECMO support. To fully assess the risk of ABI in patients on ECMO support, the integration of multi-omic data including metabolomics, proteomics, and genomics is recommended (Table 1).
Plasma biomarkers hold significant promise for the early detection and monitoring of ABI in ECMO patients. While research in other applications has been growing, there is limited data on biomarker applications in ECMO patients. The nascent stage of this research field, particularly in the context of ECMO, necessitates a comprehensive exploration and validation of these biomarkers. While some biomarkers like GFAP and NfL have shown promise, a vast array of other potential candidates remains unexplored in ECMO patients. Further research is needed to validate their clinical utility and to establish reliable biomarker panels tailored to specific patient populations and brain injury types. Future studies should focus on larger, well-defined ECMO cohorts (ECMO indications and cannulation strategy), understanding the dynamic kinetics and threshold for ECMO-associated ABI, and adopt standardized methodologies. By advancing our understanding of the complex interactions between brain injury, ECMO, and plasma biomarkers, clinicians will be better equipped to identify patients at risk, monitor their progress, and improve patient outcomes.
▪ A comprehensive evaluation and critical appraisal of a range of biomarkers including neuron-specific enolase, glial fibrillary acidic protein (GFAP), inflammatory cytokines, microRNA, tau protein, neurofilament light chain (NfL), mitochondrial DNA, cell-free DNA, lipidomics, and metabolomics have been outlined, revealing their potential in predicting acute brain injury and neurological outcomes in extracorporeal membrane oxygenation (ECMO) patients.
▪ GFAP and NfL are biomarkers that offer the most promise for clinical utility.
▪ There is a need for further multicenter research to develop standard thresholds to optimize the application of these biomarkers in patients on ECMO support.


No potential conflict of interest relevant to this article was reported.


SMC was supported by National Heart, Lung, and Blood Institute (1K23HL157610). AMG was supported by National Institute of Neurological Disorders and Stroke (5K23NS121628).




Conceptualization: SK, SMC. Funding acquisition: SMC. Writing–original draft: all authors. Writing–review & editing: all authors.

Figure 1.
Standardized neuromonitoring protocol. ECMO: extracorporeal membrane oxygenation; NIRS: near-infrared spectroscopy; EEG: electroencephalogram; GCS: Glasgow Coma Scale; CT: computed tomography; mRS: modified Rankin Scale; MRI: magnetic resonance imaging. Adapted and Retrieved from Ong CS, et al. J Thorac Cardiovasc Surg 2023;165:2104-10 [4], to which biomarkers could be supplementary once more studied in ECMO.
Figure 2.
(A) The number of publications over time looking at “biomarker,” “surrogate marker,” or “surrogate endpoint.” (B) The number of publications over time looking at “biomarker”, “surrogate marker” or “surrogate endpoint” and extracorporeal membrane oxygenation.
Table 1.
Summary of biomarkers promising for use in ECMO neuromonitoring
Serum biomarker Physiology Peak post-ABI and first elevation time Previous use Limitation
Tau Cytoskeleton stabilization in unmyelinated axons in CNS 168 hr peak, elevation post 6 hr Prediction of GCS following TBI, correlation with AIS severity Studies examining utility in adult ECMO limited by sample size
GFAP Astrocyte cytoskeletal integrity and astrogliosis 48–72 hr peak, increase after up to 2 hours GFAP increased in pediatric ECMO patients who experienced brain injury. Studies examining utility in adult ECMO limited by sample size
NfL Axonal cytoskeleton component Condition-dependent peak, increase in several hours after ABI NfL increased after hypoxic brain injury in cardiac arrest, predictor of abnormal head CT in TBI. Studies examining utility in adult ECMO limited by sample size
NSE Paracrine regulator of glia and neurons expressed in astrocytes and oligodendrocytes 48 hr peak, increase in several hours NSE is predictive of mortality and poor neurological outcomes in TBI. Levels can be falsely elevated by the presence of hemolysis.
mt-DNA Released upon cellular injury, promoting an inflammatory response Within 6 hr, uncertain peak, but increase still present after 48 hr Increased mt-DNA copy number in patients following both focal and diffuse TBI No prior studies examining utility in ECMO
miRNA Regulate cellular pathways, including those involved in neuronal repair Peak 2–7 days after TBI Specially selected miRNA profiles correlate with severity of ARDS in ECMO patients No prior studies examining neurological injury prediction in ECMO
CfDNA Double stranded DNA fragments released into the bloodstream during cellular apoptosis or necrosis High inter-patient variability, 2 hr to 1 wk peak time CfDNA levels have been shown to correlate with severity of TBI. No prior studies examining neurological injury prediction in ECMO
Inflammatory cytokines Small, secreted molecules controlling cell-cell interaction in the setting of inflammation regulation Variable by cytokine Elevations in pediatric ECMO population in those with head injury, elevations in adults associated with poor ECMO survival No prior studies on neurological injury prediction in adult ECMO population
OSM inflammatory cytokine in the IL-6 family that is expressed in neurons, astrocytes, glia, and immune cells Biphasic peak Elevated OSM associated with poor survival after VV-ECMO Conflicting evidence: elevation associated with poor survival in VV-ECMO, improved recovery in murine model
Lipidomics/metabolomics Cellular injury results in derangements in processes such as oxidative phosphorylation Variable by lipid/carbohydrate derivative Lipidome panels achieved 85% accuracy at differentiating rats having undergone TBI from control rats. No previous studies in ECMO

ECMO: extracorporeal membrane oxygenation; ABI: acute brain injury; CNS: central nervous system, GCS: Glasgow Coma Scale; TBI: traumatic brain injury; AIS: acute ischemic stroke; GFAP: glial fibrillary acidic protein; NfL: neurofilament light chain; CT: computed tomography; NSE: neuron-specific enolase; mt-DNA: mitochondrial DNA; miRNA: micro-RNA; ARDS: acute respiratory distress syndrome; CfDNA: cell-free DNA; OSM: Oncostatin M; IL: interleukin; VV: venovenous.

  • 1. Lang NW, Schwihla I, Weihs V, Kasparek M, Joestl J, Hajdu S, et al. Survival rate and outcome of extracorporeal life support (ECLS) for treatment of acute cardiorespiratory failure in trauma patients. Sci Rep 2019;9:12902. ArticlePubMedPMCPDF
  • 2. Migdady I, Rice C, Deshpande A, Hernandez AV, Price C, Whitman GJ, et al. Brain injury and neurologic outcome in patients undergoing extracorporeal cardiopulmonary resuscitation: a systematic review and meta-analysis. Crit Care Med 2020;48:e611-9.ArticlePubMed
  • 3. Cho SM, Ziai W, Mayasi Y, Gusdon AM, Creed J, Sharrock M, et al. Noninvasive neurological monitoring in extracorporeal membrane oxygenation. ASAIO J 2020;66:388-93.ArticlePubMed
  • 4. Ong CS, Etchill E, Dong J, Shou BL, Shelley L, Giuliano K, et al. Neuromonitoring detects brain injury in patients receiving extracorporeal membrane oxygenation support. J Thorac Cardiovasc Surg 2023;165:2104-10.ArticlePubMed
  • 5. Cho SM, Wilcox C, Keller S, Acton M, Rando H, Etchill E, et al. Assessing the SAfety and FEasibility of bedside portable low-field brain Magnetic Resonance Imaging in patients on ECMO (SAFE-MRI ECMO study): study protocol and first case series experience. Crit Care 2022;26:119. ArticlePubMedPMCPDF
  • 6. Kim YJ, Kim YH, Youn CS, Cho IS, Kim SJ, Wee JH, et al. Different neuroprognostication thresholds of neuron-specific enolase in shockable and non-shockable out-of-hospital cardiac arrest: a prospective multicenter observational study in Korea (the KORHN-PRO registry). Crit Care 2023;27:313. ArticlePubMedPMCPDF
  • 7. Gul SS, Huesgen KW, Wang KK, Mark K, Tyndall JA. Prognostic utility of neuroinjury biomarkers in post out-of-hospital cardiac arrest (OHCA) patient management. Med Hypotheses 2017;105:34-47.ArticlePubMed
  • 8. Lee D, Cho Y, Ko Y, Heo NH, Kang HG, Han S. Neuron-specific enolase level as a predictor of neurological outcome in near-hanging patients: a retrospective multicenter study. PLoS One 2021;16:e0246898. ArticlePubMedPMC
  • 9. Bembea MM, Rizkalla N, Freedy J, Barasch N, Vaidya D, Pronovost PJ, et al. Plasma biomarkers of brain injury as diagnostic tools and outcome predictors after extracorporeal membrane oxygenation. Crit Care Med 2015;43:2202-11.ArticlePubMed
  • 10. Floerchinger B, Philipp A, Camboni D, Foltan M, Lunz D, Lubnow M, et al. NSE serum levels in extracorporeal life support patients: relevance for neurological outcome? Resuscitation 2017;121:166-71.ArticlePubMed
  • 11. Ramont L, Thoannes H, Volondat A, Chastang F, Millet MC, Maquart FX. Effects of hemolysis and storage condition on neuron-specific enolase (NSE) in cerebrospinal fluid and serum: implications in clinical practice. Clin Chem Lab Med 2005;43:1215-7.ArticlePubMed
  • 12. Hol EM, Pekny M. Glial fibrillary acidic protein (GFAP) and the astrocyte intermediate filament system in diseases of the central nervous system. Curr Opin Cell Biol 2015;32:121-30.ArticlePubMed
  • 13. Niebrój-Dobosz I, Rafałowska J, Lukasiuk M, Pfeffer A, Mossakowski MJ. Immunochemical analysis of some proteins in cerebrospinal fluid and serum of patients with ischemic strokes. Folia Neuropathol 1994;32:129-37.PubMed
  • 14. Herrmann M, Vos P, Wunderlich MT, de Bruijn CH, Lamers KJ. Release of glial tissue-specific proteins after acute stroke: a comparative analysis of serum concentrations of protein S-100B and glial fibrillary acidic protein. Stroke 2000;31:2670-7.ArticlePubMed
  • 15. Dvorak F, Haberer I, Sitzer M, Foerch C. Characterisation of the diagnostic window of serum glial fibrillary acidic protein for the differentiation of intracerebral haemorrhage and ischaemic stroke. Cerebrovasc Dis 2009;27:37-41.ArticlePubMedPDF
  • 16. Kaneko T, Kasaoka S, Miyauchi T, Fujita M, Oda Y, Tsuruta R, et al. Serum glial fibrillary acidic protein as a predictive biomarker of neurological outcome after cardiac arrest. Resuscitation 2009;80:790-4.ArticlePubMed
  • 17. Hayashida H, Kaneko T, Kasaoka S, Oshima C, Miyauchi T, Fujita M, et al. Comparison of the predictability of neurological outcome by serum procalcitonin and glial fibrillary acidic protein in postcardiac-arrest patients. Neurocrit Care 2010;12:252-7.ArticlePubMedPDF
  • 18. Wiesmann M, Steinmeier E, Magerkurth O, Linn J, Gottmann D, Missler U. Outcome prediction in traumatic brain injury: comparison of neurological status, CT findings, and blood levels of S100B and GFAP. Acta Neurol Scand 2010;121:178-85.ArticlePubMed
  • 19. Pelinka LE, Kroepfl A, Schmidhammer R, Krenn M, Buchinger W, Redl H, et al. Glial fibrillary acidic protein in serum after traumatic brain injury and multiple trauma. J Trauma 2004;57:1006-12.ArticlePubMed
  • 20. Pelinka LE, Kroepfl A, Leixnering M, Buchinger W, Raabe A, Redl H. GFAP versus S100B in serum after traumatic brain injury: relationship to brain damage and outcome. J Neurotrauma 2004;21:1553-61.ArticlePubMed
  • 21. Nylén K, Ost M, Csajbok LZ, Nilsson I, Blennow K, Nellgård B, et al. Increased serum-GFAP in patients with severe traumatic brain injury is related to outcome. J Neurol Sci 2006;240:85-91.ArticlePubMed
  • 22. Fraser DD, Close TE, Rose KL, Ward R, Mehl M, Farrell C, et al. Severe traumatic brain injury in children elevates glial fibrillary acidic protein in cerebrospinal fluid and serum. Pediatr Crit Care Med 2011;12:319-24.ArticlePubMed
  • 23. Amalia L. Glial fibrillary acidic protein (GFAP): neuroinflammation biomarker in acute ischemic stroke. J Inflamm Res 2021;14:7501-6.ArticlePubMedPMCPDF
  • 24. Kapoor S, Ahmad SA, Muquit S, Gusdon A, Khanduja S, Ziai W, et al. Brain injury plasma biomarkers in patients on VA-ECMO: a pilot prospective observational study. [Preprint]. 2023 [cited 2023 Nov 20]. Available from:
  • 25. Avila J, Lucas JJ, Perez M, Hernandez F. Role of tau protein in both physiological and pathological conditions. Physiol Rev 2004;84:361-84.ArticlePubMed
  • 26. Gao Y, Tan L, Yu JT, Tan L. Tau in Alzheimer’s disease: mechanisms and therapeutic strategies. Curr Alzheimer Res 2018;15:283-300.ArticlePubMedPDF
  • 27. Zuckerman SL, Brett BL, Jeckell A, Yengo-Kahn AM, Solomon GS. Chronic traumatic encephalopathy and neurodegeneration in contact sports and American football. J Alzheimers Dis 2018;66:37-55.ArticlePubMed
  • 28. Magnoni S, Esparza TJ, Conte V, Carbonara M, Carrabba G, Holtzman DM, et al. Tau elevations in the brain extracellular space correlate with reduced amyloid-β levels and predict adverse clinical outcomes after severe traumatic brain injury. Brain 2012;135(Pt 4):1268-80.ArticlePubMedPMC
  • 29. De Vos A, Bjerke M, Brouns R, De Roeck N, Jacobs D, Van den Abbeele L, et al. Neurogranin and tau in cerebrospinal fluid and plasma of patients with acute ischemic stroke. BMC Neurol 2017;17:170. ArticlePubMedPMCPDF
  • 30. Mattsson N, Zetterberg H, Nielsen N, Blennow K, Dankiewicz J, Friberg H, et al. Serum tau and neurological outcome in cardiac arrest. Ann Neurol 2017;82:665-75.ArticlePubMedPMCPDF
  • 31. Gabbita SP, Scheff SW, Menard RM, Roberts K, Fugaccia I, Zemlan FP. Cleaved-tau: a biomarker of neuronal damage after traumatic brain injury. J Neurotrauma 2005;22:83-94.ArticlePubMed
  • 32. Gaetani L, Blennow K, Calabresi P, Di Filippo M, Parnetti L, Zetterberg H. Neurofilament light chain as a biomarker in neurological disorders. J Neurol Neurosurg Psychiatry 2019;90:870-81.ArticlePubMed
  • 33. Gafson AR, Barthélemy NR, Bomont P, Carare RO, Durham HD, Julien JP, et al. Neurofilaments: neurobiological foundations for biomarker applications. Brain 2020;143:1975-98.ArticlePubMedPMCPDF
  • 34. Thebault S, Booth RA, Freedman MS. Blood neurofilament light chain: the neurologist’s troponin? Biomedicines 2020;8:523. ArticlePubMedPMC
  • 35. Moseby-Knappe M, Mattsson N, Nielsen N, Zetterberg H, Blennow K, Dankiewicz J, et al. Serum neurofilament light chain for prognosis of outcome after cardiac arrest. JAMA Neurol 2019;76:64-71.ArticlePubMedPMC
  • 36. Wihersaari L, Reinikainen M, Furlan R, Mandelli A, Vaahersalo J, Kurola J, et al. Neurofilament light compared to neuron-specific enolase as a predictor of unfavourable outcome after out-of-hospital cardiac arrest. Resuscitation 2022;174:1-8.ArticlePubMed
  • 37. Levin H, Lybeck A, Frigyesi A, Arctaedius I, Thorgeirsdóttir B, Annborn M, et al. Plasma neurofilament light is a predictor of neurological outcome 12 h after cardiac arrest. Crit Care 2023;27:74. ArticlePubMedPMCPDF
  • 38. Martucci G, Arcadipane A, Tuzzolino F, Occhipinti G, Panarello G, Carcione C, et al. Identification of a circulating miRNA signature to stratify acute respiratory distress syndrome patients. J Pers Med 2020;11:15. ArticlePubMedPMC
  • 39. Zhang JM, An J. Cytokines, inflammation, and pain. Int Anesthesiol Clin 2007;45:27-37.ArticlePubMedPMC
  • 40. Datzmann T, Träger K. Extracorporeal membrane oxygenation and cytokine adsorption. J Thorac Dis 2018;10(Suppl 5):S653-60.ArticlePubMedPMC
  • 41. Caprarola SD, Ng DK, Carroll MK, Tekes A, Felling RJ, Salorio CF, et al. Pediatric ECMO: unfavorable outcomes are associated with inflammation and endothelial activation. Pediatr Res 2022;92:549-56.ArticlePubMedPMCPDF
  • 42. Burrell AJC, Lubnow M, Enger TB, Nanjayya VB, Philipp A, Malfertheiner MV, et al. The impact of venovenous extracorporeal membrane oxygenation on cytokine levels in patients with severe acute respiratory distress syndrome: a prospective, observational study. Crit Care Resusc 2017;19(Suppl 1):37-44.
  • 43. Setiadi H, El Banayosy AM, Koerner MM, Maybauer MO, Harper MD, Horstmanshof DA, et al. Oncostatin M as a biomarker to predict the outcome of VV ECMO supported patients with acute pulmonary failure. J Heart Lung Transplant 2020;39(4 Supplement):S390. Article
  • 44. Guo S, Li ZZ, Gong J, Xiang M, Zhang P, Zhao GN, et al. Oncostatin M confers neuroprotection against ischemic stroke. J Neurosci 2015;35:12047-62.ArticlePubMedPMC
  • 45. Knight D. Oncostatin M. In: Laurent GJ, Shapiro SD, editors. Encyclopedia of respiratory medicine. 2006; Academic Press. p. 254-8.
  • 46. Harrington JS, Huh JW, Schenck EJ, Nakahira K, Siempos II, Choi AM. Circulating mitochondrial DNA as predictor of mortality in critically ill patients: a systematic review of clinical studies. Chest 2019;156:1120-36.ArticlePubMedPMC
  • 47. Shimada K, Crother TR, Karlin J, Dagvadorj J, Chiba N, Chen S, et al. Oxidized mitochondrial DNA activates the NLRP3 inflammasome during apoptosis. Immunity 2012;36:401-14.ArticlePubMedPMC
  • 48. Nakahira K, Hisata S, Choi AM. The roles of mitochondrial damage-associated molecular patterns in diseases. Antioxid Redox Signal 2015;23:1329-50.ArticlePubMedPMC
  • 49. Mao JY, Li DK, Zhang HM, Wang XT, Liu DW. Plasma mitochondrial DNA levels are associated with acute lung injury and mortality in septic patients. BMC Pulm Med 2021;21:66. ArticlePubMedPMCPDF
  • 50. Puskarich MA, Shapiro NI, Trzeciak S, Kline JA, Jones AE. Plasma levels of mitochondrial DNA in patients presenting to the emergency department with sepsis. Shock 2012;38:337-40.ArticlePubMedPMC
  • 51. Fan JX, Zeng L, Chen L, Li Y, Pu H, Shen J, et al. The hidden link between plasma mitochondrial DNA level and cardiac dysfunction after cardiopulmonary bypass. Research Square [Preprint]. 2020 [cited 2023 Nov 20]. Available from:
  • 52. Bynum J, Meyer A, Darlington DN, McIntosh C, Peltier G, Taylor AS, et al. Increased mitochondrial DNA in an ECMO model is associated with loss of platelet function. Blood 2017;130(Suppl 1):1126.
  • 53. Tsai NW, Lin TK, Chen SD, Chang WN, Wang HC, Yang TM, et al. The value of serial plasma nuclear and mitochondrial DNA levels in patients with acute ischemic stroke. Clin Chim Acta 2011;412:476-9.ArticlePubMed
  • 54. Corcoran RB, Chabner BA. Application of cell-free DNA analysis to cancer treatment. N Engl J Med 2018;379:1754-65.ArticlePubMed
  • 55. Bustamante A, Mancha F, Macher HC, García-Berrocoso T, Giralt D, Ribó M, et al. Circulating cell-free DNA is a predictor of short-term neurological outcome in stroke patients treated with intravenous thrombolysis. J Circ Biomark 2016;5:1849454416668791. ArticlePubMedPMCPDF
  • 56. Vajpeyee A, Wijatmiko T, Vajpeyee M, Taywade O, Pandey S, Chauhan PS. Clinical usefulness of cell-free DNA as a prognostic marker in acute ischemic stroke. Neurologist 2020;25:11-3.ArticlePubMed
  • 57. Rainer TH, Wong LK, Lam W, Yuen E, Lam NY, Metreweli C, et al. Prognostic use of circulating plasma nucleic acid concentrations in patients with acute stroke. Clin Chem 2003;49:562-9.ArticlePubMedPDF
  • 58. Wang HC, Lin YJ, Lin WC, Ho JT, Chen WF, Chang WN, et al. The value of serial plasma nuclear and mitochondrial DNA levels in acute spontaneous intra-cerebral haemorrhage. Eur J Neurol 2012;19:1532-8.PubMed
  • 59. Hogan SR, Phan JH, Alvarado-Velez M, Wang MD, Bellamkonda RV, Fernández FM, et al. Discovery of lipidome alterations following traumatic brain injury via high-resolution metabolomics. J Proteome Res 2018;17:2131-43.ArticlePubMedPMC
  • 60. Gier EC, Pulliam AN, Gaul DA, Moore SG, LaPlaca MC, Fernández FM. Lipidome alterations following mild traumatic brain injury in the rat. Metabolites 2022;12:150. ArticlePubMedPMC
  • 61. Thomas I, Dickens AM, Posti JP, Czeiter E, Duberg D, Sinioja T, et al. Serum metabolome associated with severity of acute traumatic brain injury. Nat Commun 2022;13:2545. PubMedPMC
  • 62. Gusdon AM, Savarraj JP, Redell JB, Paz A, Hinds S, Burkett A, et al. Lysophospholipids are associated with outcomes in hospitalized patients with mild traumatic brain injury. J Neurotrauma 2023 Oct 20 [Epub].

Figure & Data



    Citations to this article as recorded by  

      • PubReader PubReader
      • ePub LinkePub Link
      • Cite
        export Copy
        Download Citation
        Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

        • RIS — For EndNote, ProCite, RefWorks, and most other reference management software
        • BibTeX — For JabRef, BibDesk, and other BibTeX-specific software
        • Citation for the content below
        Plasma biomarkers for brain injury in extracorporeal membrane oxygenation
        Acute Crit Care. 2023;38(4):389-398.   Published online November 29, 2023
      • XML DownloadXML Download
      Related articles

      ACC : Acute and Critical Care