Secondary Logo

Journal Logo

Original Articles: Hepatology

Hepatic Encephalopathy in Children With Acute Liver Failure: Utility of Serum Neuromarkers

Toney, Nicole A.; Bell, Michael J.; Belle, Steven H.; Hardison, Regina M.; Rodriguez-Baez, Norberto; Loomes, Kathleen M.§; Vodovotz, Yoram||; Zamora, Ruben||; Squires, Robert H.; for the Pediatric Acute Liver Failure Study Group

Author Information
Journal of Pediatric Gastroenterology and Nutrition: July 2019 - Volume 69 - Issue 1 - p 108-115
doi: 10.1097/MPG.0000000000002351


What Is Known/What Is New

What Is Known

  • Children with acute liver failure and hepatic encephalopathy scores that are high or progress are associated with poor outcomes.
  • Children who were alive 1 year following their episode of acute liver failure experience significant impairments in neurocognitive skills and quality of life.

What Is New

  • Serum neuromarkers (neuron-specific enolase, S100β, and myelin basic protein) are detectable in children with pediatric acute liver failure.
  • Serum S100β and IL-6, an inflammatory marker, are associated with hepatic encephalopathy in children with acute liver failure.
  • Measuring these markers may assist in assessing neurological injury in pediatric acute liver failure, impacting clinical decisions.

Pediatric acute liver failure (PALF) is an acute form of liver injury manifested by a rapid onset of symptoms, coagulopathy, and encephalopathy. Clinical outcomes of PALF include recovery of liver function after extensive medical treatment(s), survival after successful liver transplant, or death—with or without liver transplant (1–3). In a recent observational study of PALF, we found that 53% recovered, 32% underwent transplantation, and 14% died (4). Furthermore, we have demonstrated that survivors of PALF have decreased motor abilities, attention, executive function, and quality of life compared to age-matched controls (5). Thus, although rare, PALF has high mortality and morbidity for an individual child and a large burden on families and for the public health.

Hepatic encephalopathy (HE) is often clinically evident in PALF, although its pathophysiology is not fully understood. Presence of HE influences decision making for medical therapies and decisions to list for and proceed with liver transplantation (6). Clinical assessment of HE currently relies upon the experience of the individual examining the patient and significant variability in observation and manifestation of HE may exist. Therefore, objective tools to assess HE would be welcomed.

Serum neuromarkers are structural, brain-related proteins measurable in serum following neurological injury. Individual neuromarkers are derived from specialized cell types within the brain with neuron-specific enolase (NSE) present in neurons, S100β present in astrocytes, and myelin basic protein (MBP) in oligodendrocytes. Recent evidence determined that brain-derived proteins from within the neurovascular unit can escape into the peripheral circulation despite an intact blood-brain barrier (7–9). Many neuromarkers have been demonstrated to be increased after brain injuries such as hypoxic/ischemic injury after cardiac arrest, traumatic brain injury, stroke, and others (10–17). Most have been used to either diagnose an injury or to provide some evidence for injury severity—yielding a more objective assessment of neurological injury. Although reports of neurological markers to detect brain injury associated with intracranial hypertension in patients with traumatic brain injury (18,19) are noted, similar studies have not been performed in children with acute liver failure (ALF).

In this exploratory study, we intend to test the hypotheses that serum neuromarkers (NSE, S100β, and MBP) are detectable in children with PALF and that these neuromarkers are associated with the degree of HE. In addition, we will use circulating inflammatory mediators previously reported (20) to determine whether an association between inflammatory markers and encephalopathy is present in PALF participants.


Patient Enrollment

The PALF study is a multicenter, observational cohort study of infants and children with acute liver failure. All medical decisions for the care of children, including medical, surgical, and transplant decisions, were made by the clinical teams of each institution. Approval from the institutional review board at each participating site was secured. Informed consent for the child to participate in the PALF study was obtained from the parent/caregiver (and assent from study subjects, as appropriate). Participants for this analysis were enrolled between May 2012 and December 2014 by 12 participating centers. Consent included permission to collect and store biospecimens for future research. This report represents one such investigation with measurement of neuromarkers and inflammatory markers from stored specimens.

Patients from birth through 18 years of age were eligible for enrollment if they met the following entry criteria for the PALF study: no known evidence of chronic liver disease, evidence of acute liver injury, and hepatic-based coagulopathy not corrected by vitamin K with the following parameters: prothrombin time ≥15 seconds or international normalized ratio (INR) ≥1.5 in the presence of clinical HE or a prothrombin time ≥20 seconds or INR ≥2.0 regardless of the presence or absence of clinical HE (4). For this ancillary study, an additional inclusion criterion was the availability of a research serum sample for analyzing neurological and other markers with concomitant assessment of HE on at least one of the days a sample was available. Exclusion criteria for the PALF study were known chronic underlying liver disease; multiorgan system failure following heart surgery or Extracorporeal Membrane Oxygenation; solid organ or bone marrow transplantation; acute trauma; previous enrollment in the PALF study.

Study Procedures

Data for the PALF study were collected daily from enrollment until discharge with the native liver, liver transplant, or death. Data extracted from the medical records included diagnostic studies, morning laboratory tests obtained as part of standard care to assess the child's clinical condition, and documentation of medical/surgical therapies administered to the child. All children were assessed for HE with a score of 0 to 4, with 0 corresponding to no encephalopathy, using the Whittington scale (21) for infants and children younger than 3 years of age and the West Haven criteria (22) for those 3 to 18 years. The study investigator assessed participants each day at the earliest morning encounter to approximate clinical and research data with bedside clinical assessment.

Measurements of Markers and Data Analysis

Weight-based daily and monthly blood volume restrictions for obtaining additional blood samples for research purposes were observed. If allowable, a daily serum sample for research purposes was collected at the time of the first morning blood draw, stored locally at −80°C, and then batch-shipped for long-term storage at the National Institute of Diabetes and Digestive and Kidney Diseases funded Biorepository. Serum samples from three time points that met criteria for analysis were shipped to the investigators. Given the variability of the interval among participants and the dynamic clinical course of PALF, preference for serum selection of the 3 qualifying study days was for one to be nearest to enrollment, 1 day nearest the outcome, and the third day nearest the middle of the interval for the participant. Serum concentrations of NSE, S100β, and MBP were measured using commercially available enzyme-linked immunosorbent assay kits. In general, serum was applied to 96-well kits along with samples of known concentrations and controls (both negative and positive). Sandwich enzyme-linked immunosorbent assays were performed using biotin-labeled antibodies for primary antigen detection. Optical density was measured at 450 nm by a microplate reader (Thermomax Microplate Reader, Molecular Devices, Sunnyvale, CA), and data were analyzed with Soft Max Pro software, version 5.0.1 (Molecular Devices). Standard curves were determined from known standards and sample neuromarker concentrations were calculated.

Cytokine measurements were performed as previously reported (20). Briefly, the human inflammatory MILLIPLEX MAP Human Cytokine/Chemokine Panel-Premixed 24 Plex (Millipore, Billerica, MA and a Luminex 100 IS apparatus (Luminex, Austin, TX) were used to measure serum levels of Interleukin (IL)-1β, IL-1 receptor antagonist (IL-1RA), IL-2, soluble IL-2 receptor-α (sIL-2Rα), IL-4, IL-5, IL-6, IL-7, IL-8 (CCL8), IL-10, IL-13, IL-15, IL-17, interferon-γ, interferon-γ inducible protein-10 (CXCL10), monokine induced by gamma interferon (MIG; CXCL9), macrophage inflammatory protein-1 (CCL2), granulocyte-macrophage colony stimulating factor, eotaxin (CCL11), and tumor necrosis factor.

For all but 1 participant, serum samples were available for 3 days; 1 patient had 4 samples available. For a participant to be included in this analysis they had to have at least 1 serum sample and an evaluation of HE on the same day. Participants were followed for the interval between enrollment and the outcome of hospital discharge, liver transplant, or death. Laboratory personnel (N.A.T. and M.J.B.) were masked with respect to HE results and all PALF study staff were masked with respect to neuromarker concentrations until they were generated by the lab personnel. Because of the observed distribution of the HE assessments, and the difficulty distinguishing HE grade 1 from grade 0, HE grading was dichotomized to “No HE” (grades 0 and 1) and “HE” (grades 2–4). If the child's encephalopathy grade could not be assessed because the child was on a respirator, “HE” (grades 2–4) was assigned a priori. We chose this approach with the assumption that participants on the respirator at the time clinical assessment of HE was to be determined were highly likely to have HE that was at least stage 2. If unassessable for other reasons, such as use of sedative medications, HE was considered missing.

Descriptive statistics are shown using frequencies and percentages for categorical data and medians, 25th and 75th percentiles, minima, and maxima for continuous data. Differences in distributions between participants in the PALF cohort who were included in the analysis and those who were not were tested using the Wilcoxon test for continuous data or the Pearson chi-square test of association for categorical data. If the expected value in any cell for categorical data was <5, an exact test was used. Gray test was used to test whether the cumulative incidences of death and transplantation differed significantly between those included in the analyses and the others in the PALF cohort. The proportional hazards assumption for both death and liver transplantation were tested by assessing the statistical significance of the interaction between time and the indicator variable for inclusion in the analysis.

Due to extreme values, when examining the association of HE with neuromarkers, the neuromarkers were first transformed using log2. They were individually examined for association with HE using a simple generalized linear mixed model with a logit link (binomial) and random intercept to estimate the odds ratio (OR) for encephalopathy. These models were used to account for the repeated measures of the neuromarkers in individual participants. The model assumes that data are missing at random. Overdispersion was tested using the ratio of the Pearson chi-square test and degrees of freedom. The logit of the probability of HE was plotted against each neuromarker. When there was evidence that the assumption of a linear relationship was not met, then quadratic and, in some cases, higher-order polynomial terms were included in the model. Multiple generalized linear mixed models were used to examine the association of HE and each neuromarker, adjusting separately for age at sample draw, sex, and race. All available data were used in this analysis for the 82 participants with at least 1 neuromarker measure and encephalopathy grade on the same day. Due to the multiple comparisons (3 neuromarkers, 28 cytokines, and other analytes), the probability of incorrectly finding a statistically significant result (type I error) is inflated. Therefore, to reduce the probability of finding a significant association due to the many comparisons performed, Holm stepdown method was used. An adjusted P value of 0.05 was used to determine statistical significance. ORs for encephalopathy per unit change in log2 biomarker (ie, doubling the value) are presented. All statistical analyses were performed with SAS 9.4 software (SAS Institute, Cary, NC) and figure was drawn using R 3.4.0 (R Foundation for Statistical Computing, Vienna, Austria).


Of the 158 participants in this phase of PALF, 82 had serum samples available for analysis on the same day HE assessment was recorded. The median age was 2.9 years, ranging from infants (6 days old) through 17.8 years (25th–75th percentiles 0.1–13.7 years). Most children were boys (63%) and nearly 3/4 were Caucasian (74%). At the time of enrollment, 11% of subjects had HE grade >1. Demographic, clinical, and outcome information for those meeting criteria for the neuromarker study and those not included are provided in Table 1. There was no evidence of nonproportionality of hazards for either death or liver transplantation. There were no significant differences between those eligible for the neuromarker study and those who were not.

Comparison between neuromarker analysis subset participants and the pediatric acute liver failure study cohort

The etiology of acute liver failure and laboratory analysis on study entry are also provided in Table 1. Consistent with previous findings (4), an indeterminate diagnosis (29%) was the most common final diagnosis, with acetaminophen intoxication being the second most common (11%). Children were quite icteric at study entry (total bilirubin; 7.6 mg/dL [3.8–15.5], median [25th–75th percentile]), but largely had normal renal function (Cr; 0.4 mg/dL [0.2–0.7]). alanine aminotransferase (1598 IU/L [500–3817]) and INR (2.7 [2.2–3.9]) values are consistent with severe hepatic injury and diminished synthetic function of the children at the time of study entry.

Of the 82 participants in the study, 81 had 3 serum samples available for analysis and 1 child had 4 samples that were analyzed yielding a total of 247 samples. There were 80 samples (32% of the total) obtained on the day of, or the day after, study entry and 36 samples (15%) were obtained 10 or more days after study entry. HE was present during the study period in 37 of 82 (45%) of participants. The distribution of HE scores on the day of the serum sample is shown in Table 2. For 20 (8%) time points there were missing HE scores, either because they were not collected or because the HE could not be assessed, most often due to concurrent administration of sedatives or barbiturates. Of those that were measured, 161 (71%) time points were “no HE,” whereas 66 (29%) scores corresponded to “HE” group.

Distribution of hepatic encephalopathy scores with serum sampling

Table 3 segregates those who were never, ever, or always on the ventilator at a time point when HE assessment and sample collection were paired to examine age and clinical characteristics of participants. Participants who were always on the ventilator were younger and more likely to have a viral etiology and die, whereas those never on the ventilator were more likely to have an indeterminate diagnosis and receive a liver transplant.

Participant ventilator status for participants with hepatic encephalopathy at time of neuromarker sample hepatic encephalopathy is defined as either hepatic encephalopathy 2, 3, 4, or nonassessable hepatic encephalopathy and on a ventilator

Progression of HE is associated with a higher incidence of liver transplantation and death (23). Clinical assessment of HE progression can only occur in individuals who are neither sedated nor ventilated. HE, however, cannot be assessed for participants on the ventilator. Therefore, we assumed those on the ventilator were highly likely to have HE grade 2–4. As noted in Table 3, 10 of 18 (66%) of the participants who were never ventilated had clinical HE grade 0–1 at study entry, but progressed to grade 2–4 and this was associated with a high frequency of liver transplantation in the never ventilated cohort. For those always ventilated at a study timepoint, 3 of 17 had HE 0–1 assessed at study entry, but all progressed to being on the ventilator and the ventilated group had a high rate of mortality. The dynamic nature of HE among PALF participants, coupled with the high mortality rate among those always on the ventilator supports our assumption to classify ventilated participants as having HE grade 2–4.

Neuromarker concentrations at each of the 3 time points are summarized in Table 4. There was a significant relationship between S100β and HE (OR 1.16, 95% confidence interval [1.03–1.29], adjusted P = 0.04) (Table 5). With a doubling of S100β levels there was a 16% increased odds of having HE (grades 2–4). There were no significant associations between HE and either MBP or NSE. Cytokine concentrations from each of the 3 sampling time points are summarized in Supplemental Table 1 (Supplemental Digital Content, Only IL-6 was significantly associated (P = 0.003) with HE after adjusting for multiple comparisons. The relationship was nonlinear such that the odds of encephalopathy was 35% higher ([1.18–1.54]) when the level of IL-6 doubled from the median value of 13 to 26 pg/mL, increasing to 71% higher ([1.34–2.17]) when IL-6 level doubled from the mean value of 127 to 254 pg/mL).

Neuromarker concentrations over study time periods
Logistic modeling of hepatic encephalopathy with neuromarkers


In this exploratory analysis from a multicenter, observational PALF study, we are the first to test the utility of measuring serum neuromarkers specific to neurons, astrocytes, and oligodendrocytes during the acute period of illness. We demonstrated that HE was significantly associated with both the astrocyte marker S100β and the proinflammatory cytokine IL-6. To our knowledge, this is the first report that has demonstrated these associations in children and may lead to improved monitoring methods in the future.

The pathophysiology of ALF, and the manifestation of HE during the course of the illness, is complex (24,25). Any number of mechanisms including alterations of cerebral blood flow/cerebrohemodynamics with intracranial hypertension, effects of metabolic derangements (including hyperammonemia among others), oxidative stress, and inflammatory cascades have been implicated in HE development. Despite these disparate mechanisms, the interaction of neurons and glia is critical as glial elements play important roles in water homeostasis, neurotransmitter metabolism and clearance of byproducts as neurons may become dysfunctional during HE. For this reason, we endeavored to categorize the neurological markers related to the main subtypes of cells within the central nervous system (neurons, astrocytes, and oligodendrocytes) and found that the astrocyte marker was associated with HE.

Clinically, HE remains a main driving force for adverse outcomes for ALF. Ciocca et al (6) demonstrated that HE and coagulopathy were associated with death or need for transplantation in 215 children from Argentina, and Ng et al (23) demonstrated that severe HE (grade III or IV) or progression of encephalopathy was associated with 21 day mortality in 769 children in the United States. In a smaller series with 19 children, Hussain et al (26) demonstrated some utility of electroencephalogram monitoring of such children as those with the moderate-to-severe abnormalities were more likely to die or require transplantation. They, however, acknowledge additional neurological assessment strategies will be useful in medical and transplant decisions.

Although this is the first study to investigate the role of neuromarkers in PALF, several reports have been published regarding the association of neuromarkers in adults with liver disease. Several decades ago, Kimura and Budka (27) demonstrated astrocyte activation in brains of patients with HE at autopsy. Shiotani et al (28) demonstrated that S100β was associated with perioperative brain edema in 13 adults undergoing liver transplantation for fulminant hepatic failure, whereas Isobe-Harima et al (29) demonstrated that S100β was increased at the onset of HE in 9 adults with hepatitis. Similar to our findings, Saleh et al (30) found that S100β was increased in 29 cirrhotic adults with stage 1 and 2 HE, whereas NSE was not changed. In the 2 largest series, Strauss et al (31) found that S100β was increased in most patients with fulminant hepatic failure (n = 35) and acute-on-chronic liver disease (n = 6), whereas NSE was increased in subjects who developed cerebral herniation. And using samples from 54 subjects in the US Acute Liver Failure Study Group, Vaquero et al (32) found S100β was increased in subjects with stage 1–2 HE who did not progress, stage 1–2 HE who progressed to a deeper encephalopathy, stage 3–4 HE who survived, and stage 3–4 HE who died or required transplantation. They, however, failed to find a difference among these 4 groups—thereby arguing that S100β is not useful in this condition. Our data from 82 children are the largest of these studies and does demonstrate an association between S100β and the presence of HE. Given that the assessment of children and infants with severe liver disease may be more difficult than adults, we believe that there may be more utility in this neuromarker in children with PALF than in these previous studies.

As stated previously, one of the goals of PALF was to derive a model of inflammation and repair that would predict the need for liver transplantation (20). The link between neuroinflammation with acute liver failure has long been postulated (33), from evidence of experimental models demonstrating that inflammation can precipitate HE (34) to evidence of microglial proliferation in patients with HE (35). In our study, we tested the relationship between selected inflammatory mediators with HE and found that only IL-6 had a significant association. Similar to our study, Tsai et al (36) found that IL-6 was associated with mild HE in 94 adults with cirrhosis, whereas Srivastava et al (37) demonstrated that both tumor necrosis factor and IL-6 were increased in 20 adults with minimal HE—along with magnetic resonance imaging findings and ammonia concentrations. Should the association between HE in PALF and elevations in S100β and IL-6 be consistently demonstrated in other studies, it is possible that use of these simple blood tests could prove useful in evaluating children with PALF and HE.

There are several limitations to the study. First, the sample size is modest, although larger than others cited above, and the sampling was done at various times during the child's illness. The study design attempted to minimize this limitation by having HE and neuromarkers measured on the same day. Although sites were instructed to draw research samples with the earliest morning clinical laboratory draw and principal investigators were instructed to obtain HE scores in their first morning assessment of the day, actual times for both were not collected. Definitions of HE scoring were available and the principal investigator was encouraged to limit the number of physicians performing the daily neurological examination; however, the scoring of HE was performed by more than 1 individual during hospitalization and assignment of scores from children with mechanical ventilation was challenging. Reliability crosscenters would be somewhat mitigated by collapsing categories as we did for this analysis. As an observational study across multiple centers, it was not possible to perform detailed neurological assessments other than the HE scoring, which is a standard metric used by all participating sites as part of their clinical practice. Furthermore, it was not possible to intervene within this clinical study to decrease sedative and/or neuromuscular blockade medications to obtain these scores in a more systematic manner. A priori scoring of HE 2–4 for all participants on a ventilator may have been incorrect for individual participants.

In conclusion, this study demonstrates the potential utility of neuromarkers in the neurological assessment of children with acute liver failure and suggests associations between neurological injury with proinflammatory cytokines that are responsible for regeneration and repair. If future studies confirm these associations, it may lead to novel ways to monitor patients with acute liver failure as they are recovering or awaiting transplantation.


Key individuals who have actively participated in the PALF studies include (by site): Current Sites, Principal Investigators and Coordinators—Robert H. Squires, MD, Kathryn Bukauskas, RN, CCRC, Madeline Schulte, RN, BSN, Clinical Research Coordinator (Children's Hospital of Pittsburgh of UPMC, Pittsburgh, Pennsylvania); Michael R. Narkewicz, MD, Michelle Hite, MA, CCRC (Children's Hospital Colorado, Aurora, Colorado); Kathleen M. Loomes, MD, Elizabeth B. Rand, MD, David Piccoli, MD, Deborah Kawchak, MS, RD, Christa Seidman, Clinical Research Coordinator (Children's Hospital of Philadelphia, Philadelphia, PA); Rene Romero, MD, Saul Karpen, MD, PhD, Liezl de la Cruz-Tracy, CCRC (Emory University, Atlanta, GA); Vicky Ng, MD, Kelsey Hunt, Clinical Research Coordinator (Hospital for Sick Children, Toronto, Ontario, Canada); Girish C. Subbarao, MD, Ann Klipsch, RN, Sarah Munson, Clinical Research Coordinator (Indiana University Riley Hospital, Indianapolis, IN); Estella M. Alonso, MD, Lisa Sorenson, PhD, Susan Kelly, RN, BSN, Katie Neighbors, MPH, CCRC (Lurie Children's Hospital of Chicago, Chicago, IL); Philip J. Rosenthal, MD, Shannon Fleck, Clinical Research Coordinator (University of California San Francisco, San Francisco, CA); Mike A. Leonis, MD, PhD, John Bucuvalas, MD, Tracie Horning, Clinical Research Coordinator (University of Cincinnati, Cincinnati, OH); Norberto Rodriguez Baez, MD, Shirley Montanye, RN, Clinical Research Coordinator, Margaret Cowie, Clinical Research Coordinator (University of Texas Southwestern, Dallas, TX); Simon P. Horslen, MD, Karen Murray, MD, Melissa Young, Clinical Research Coordinator, Heather Nielson, Clinical Research Coordinator, Jani Klein, Clinical Research Coordinator (University of Washington, Seattle, WA); David A. Rudnick, MD, PhD, Ross W. Shepherd, MD, Kathy Harris, Clinical Research Coordinator (Washington University, St. Louis, MO).

Previous Sites, Principal Investigators and Coordinators – Saul J. Karpen, MD, PhD, Alejandro De La Torre, Clinical Research Coordinator (Baylor College of Medicine, Houston, TX); Dominic Dell Olio, MD, Deirdre Kelly, MD, Carla Lloyd, Clinical Research Coordinator (Birmingham Children's Hospital, Birmingham, United Kingdom); Steven J. Lobritto, MD, Sumerah Bakhsh, MPH, Clinical Research Coordinator (Columbia University, New York, NY); Maureen Jonas, MD, Scott A. Elifoson, MD, Roshan Raza, MBBS (Harvard Medical School, Boston, MA); Kathleen B. Schwarz, MD, Wikrom W. Karnsakul, MD, Mary Kay Alford, RN, MSN, CPNP (Johns Hopkins University, Baltimore, MD); Anil Dhawan, MD, Emer Fitzpatrick, MD (King's College Hospital, London, United Kingdom); Nanda N. Kerkar, MD, Brandy Haydel, CCRC, Sreevidya Narayanappa, Clinical Research Coordinator (Mt. Sinai School of Medicine, New York, NY); M. James Lopez, MD, PhD, Victoria Shieck, RN, BSN (University of Michigan, Ann Arbor, MI).

The authors are also grateful for support from the National Institutes of Health (Edward Doo, MD, Director Liver Disease Research Program, and Averell H. Sherker, MD, Scientific Advisor, Viral Hepatitis and Liver Diseases, DDDN-NIDDK) and for assistance from members of the Data Coordinating Center at the University of Pittsburgh (directed by Steven H. Belle, PhD, MScHyg).


1. Cochran JB, Losek JD. Acute liver failure in children. Pediatr Emerg Care 2007; 23:129–135.
2. D’Agostino D, Diaz S, Sanchez MC, et al. Management and prognosis of acute liver failure in children. Curr Gastroenterol Rep 2012; 14:262–269.
3. Treem WR. Fulminant hepatic failure in children. J Pediatr Gastroenterol Nutr 2002; 35 (suppl 1):S33–S38.
4. Squires RH Jr, Shneider BL, Bucuvalas J, et al. Acute liver failure in children: the first 348 patients in the pediatric acute liver failure study group. J Pediatr 2006; 148:652–658.
5. Sorensen LG, Neighbors K, Zhang S, et al. Neuropsychological functioning and health-related quality of life: pediatric acute liver failure study group results. J Pediatr Gastroenterol Nutr 2015; 60:75–83.
6. Ciocca M, Ramonet M, Cuarterolo M, et al. Prognostic factors in paediatric acute liver failure. Arch Dis Child 2008; 93:48–51.
7. Aspelund A, Antila S, Proulx ST, et al. A dural lymphatic vascular system that drains brain interstitial fluid and macromolecules. J Exp Med 2015; 212:991–999.
8. Iliff JJ, Wang M, Liao Y, et al. A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid beta. Sci Transl Med 2012; 4:147ra111.
9. Plog BA, Dashnaw ML, Hitomi E, et al. Biomarkers of traumatic injury are transported from brain to blood via the glymphatic system. J Neurosci 2015; 35:518–526.
10. Yardan T, Cevik Y, Donderici O, et al. Elevated serum S100B protein and neuron-specific enolase levels in carbon monoxide poisoning. Am J Emerg Med 2009; 27:838–842.
11. Zurek J, Fedora M. The usefulness of S100B, NSE, GFAP, NF-H, secretagogin and Hsp70 as a predictive biomarker of outcome in children with traumatic brain injury. Acta Neurochir (Wien) 2012; 154:93–103. discussion 103.
12. Studer M, Goeggel Simonetti B, Heinks T, et al. Acute S100B in serum is associated with cognitive symptoms and memory performance 4 months after paediatric mild traumatic brain injury. Brain Inj 2015; 29:1667–1673.
13. Varrica A, Satriano A, Tettamanti G, et al. Predictors of ominous outcome in infants who undergo cardiac surgery and cardiopulmonary by-pass: S100B protein. CNS Neurol Disord Drug Targets 2015; 14:85–90.
14. Rodriguez-Rodriguez A, Egea-Guerrero JJ, Gordillo-Escobar E, et al. S100B and Neuron-Specific Enolase as mortality predictors in patients with severe traumatic brain injury. Neurol Res 2016; 38:130–137.
15. Usui A, Kato K, Murase M, et al. Neural tissue-related proteins (NSE, G0 alpha, 28-kDa calbindin-D, S100b and CK-BB) in serum and cerebrospinal fluid after cardiac arrest. J Neurol Sci 1994; 123:134–139.
16. Soares H, Bednar M, Mennti F, et al. S100 and NSE correlate with clinical outcomes in severe traumatic brain injury. J Neurotrauma 2003; 20:1080.
17. Olivecrona Z, Bobinski L, Koskinen LO. Association of ICP, CPP, CT findings and S-100B and NSE in severe traumatic head injury. Prognostic value of the biomarkers. Brain Inj 2015; 29:446–454.
18. Petzold A, Green AJ, Keir G, et al. Role of serum S100B as an early predictor of high intracranial pressure and mortality in brain injury: a pilot study. Crit Care Med 2002; 30:2705–2710.
19. Stein DM, Lindell A, Murdock KR, et al. Relationship of serum and cerebrospinal fluid biomarkers with intracranial hypertension and cerebral hypoperfusion after severe traumatic brain injury. J Trauma 2011; 70:1096–1103.
20. Zamora R, Vodovotz Y, Mi Q, et al. Data-driven modeling for precision medicine in pediatric acute liver failure. Mol Med 2016; 22:821–829.
21. Whitington PF, Alonso EM, ed DA Kelly. DA K. Fulminant hepatitis and acute liver failure. Paediatric Liver Disease. Oxford: Blackwell; 2003. 107–126.
22. Conn HO, Leevy CM, Vlahcevic ZR, et al. Comparison of lactulose and neomycin in the treatment of chronic portal-systemic encephalopathy. A double blind controlled trial. Gastroenterology 1977; 72:573–583.
23. Ng VL, Li R, Loomes KM, et al. Outcomes of children with and without hepatic encephalopathy from the pediatric acute liver failure study group. J Pediatr Gastroenterol Nutr 2016; 63:357–364.
24. Butterworth RF. Pathogenesis of hepatic encephalopathy and brain edema in acute liver failure. J Clin Exp Hepatol 2015; 5:S96–S103.
25. Wijdicks EF. Hepatic encephalopathy. N Engl J Med 2016; 375:1660–1670.
26. Hussain E, Grimason M, Goldstein J, et al. EEG abnormalities are associated with increased risk of transplant or poor outcome in children with acute liver failure. J Pediatr Gastroenterol Nutr 2014; 58:449–456.
27. Kimura T, Budka H. Glial fibrillary acidic protein and S-100 protein in human hepatic encephalopathy: immunocytochemical demonstration of dissociation of two glia-associated proteins. Acta Neuropathol 1986; 70:17–21.
28. Shiotani S, Shimada M, Soejima Y, et al. S100 beta protein: the preoperative new clinical indicator of brain damage in patients with fulminant hepatic failure. Transplant Proc 2004; 36:2713–2716.
29. Isobe-Harima Y, Terai S, Segawa M, et al. Serum S100b (astrocyte-specific protein) is a useful marker of hepatic encephalopathy in patients with fulminant hepatitis. Liver Int 2008; 28:146–147.
30. Saleh A, Kamel L, Ghali A, et al. Serum levels of astroglial S100-beta and neuron-specific enolase in hepatic encephalopathy patients. East Mediterr Health J 2007; 13:1114–1123.
31. Strauss GI, Christiansen M, Moller K, et al. S-100b and neuron-specific enolase in patients with fulminant hepatic failure. Liver Transpl 2001; 7:964–970.
32. Vaquero J, Jordano Q, Lee WM, et al. Serum protein S-100b in acute liver failure: results of the US Acute Liver Failure Study group. Liver Transpl 2003; 9:887–888.
33. Butterworth RF. Neuroinflammation in acute liver failure: mechanisms and novel therapeutic targets. Neurochem Int 2011; 59:830–836.
34. Chastre A, Belanger M, Nguyen BN, et al. Lipopolysaccharide precipitates hepatic encephalopathy and increases blood-brain barrier permeability in mice with acute liver failure. Liver Int 2014; 34:353–361.
35. Dennis CV, Sheahan PJ, Graeber MB, et al. Microglial proliferation in the brain of chronic alcoholics with hepatic encephalopathy. Metab Brain Dis 2014; 29:1027–1039.
36. Tsai CF, Chu CJ, Huang YH, et al. Detecting minimal hepatic encephalopathy in an endemic country for hepatitis B: the role of psychometrics and serum IL-6. PLoS One 2015; 10:e0128437.
37. Srivastava A, Yadav SK, Yachha SK, et al. Pro-inflammatory cytokines are raised in extrahepatic portal venous obstruction, with minimal hepatic encephalopathy. J Gastroenterol Hepatol 2011; 26:979–986.

acute liver failure; hepatic encephalopathy; inflammatory markers; neuromarkers; pediatric acute liver failure

Supplemental Digital Content

Copyright © 2019 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition