Abstract
Objectives
To identify candidate biomarkers in both plasma and cerebrospinal fluid (CSF) that are associated with neonatal encephalopathy severity measured by encephalopathy grade, seizures, brain injury by magnetic resonance imaging (MRI), and neurodevelopmental outcomes at 15–30 months.
Study design
A retrospective cohort study of plasma (N=155, day of life 0–1) and CSF (n=30, day of life 0–7) from neonates with NE and healthy term neonates (N=30, ≥36 weeks’ gestation) was conducted. We measured CNS necrosis (glial fibrillary acidic protein [GFAP], neurogranin [NRGN], Tau), inflammatory (IL-6, IL-8, IL-10), and trophic (brain-derived neurotrophic factor [BDNF], vascular endothelial growth factor [VEGF]) proteins. Clinical outcomes were Sarnat scores of encephalopathy, seizures, MRI scores, and Bayley Scales of Infant and Toddler Development III (Bayley-III) at 15–30 months.
Results
Plasma NRGN, Tau, IL-6, IL-8, and IL-10 were higher, whereas BDNF and VEGF were lower in NE versus controls. In plasma, Tau, GFAP, and NRGN were directly and BDNF inversely associated with encephalopathy grade. IL-6 was inversely related to seizures. Tau was directly related to MRI abnormalities. Tau was inversely associated with Bayley-III cognitive and motor outcomes. In CSF, NRGN was inversely associated with cognitive, motor, and language measures. GFAP, IL-6, and IL-10 were inversely related to cognitive and motor outcomes. IL-8 was inversely related to motor outcomes. CSF candidate biomarkers showed no significant relationships with encephalopathy grade, seizures, or MRI abnormalities.
Conclusions
Plasma candidate biomarkers predicted encephalopathy severity, seizures, MRI abnormalities, and neurodevelopmental outcomes at 15–30 months.
Keywords: Neurology, development, neonate, marker, brain injury
Neonatal encephalopathy is a syndrome defined by clinical features of neurological dysfunction during the first few days of life, including difficulty initiating or maintaining respiration, altered consciousness, and seizures.1,2 NE occurs in as many as 3 per 1000 livebirths.3 Common causes include hypoxic-ischemic insult leading to hypoxic-ischemic encephalopathy (HIE), perinatal infections, maternal conditions including placental abnormalities, and neonatal conditions including metabolic disorders, coagulopathies, and neonatal stroke.1,4 The only validated treatment for this syndrome remains therapeutic hypothermia with intensive supportive care.5 Despite the successful implementation of TH, approximately 29% of neonates with NE still have unfavorable outcomes of neurological disability and death.6 The modest efficacy of TH may depend on accurate assessment of injury severity for targeted intervention.7,8 Presently, current clinical methods are insufficient to identify risk, stratify severity, and monitor therapeutic efficacy which warrants further research into multidimensional approaches to assessing and monitoring evolving brain injury in neonates with NE.9,10
As a potential component of a multidimensional diagnostic/prognostic approach, biological fluid markers can provide objective, serial, and non-invasive information to identify pathological processes reflected by distinct peripheral concentrations. Biomarker assays have been extensively studied in adults to predict neurological pathologies, including traumatic brain injury and neurodegenerative diseases.11–13 However, there is limited research and clinical application of pediatric biomarkers due to the need for large sample validation studies, especially in neonatal brain injury.14
Several candidate biomarkers, including CNS-necrosis, inflammatory, and trophic proteins, were identified in adult studies on brain injury and some have been investigated in neonatal populations. Glial fibrillary acidic protein (GFAP), a CNS-specific astrocyte cytoskeletal intermediate filament protein, was associated with abnormal magnetic resonance imaging (MRI) and neurodevelopmental outcomes.15–18 Neurogranin (NRGN), a brain-specific protein kinase C substrate, has not yet been studied in neonatal populations, but has associations with adult traumatic brain injury (TBI) and neurodegeneration.11,19–22 Tau, another CNS-necrosis factor, was associated with asphyxia and NE.23–25 Inflammation plays an important role in brain injury, and increased cytokines, especially interleukins 6, 8, and 10 (IL-6, IL-8, IL-10), have been associated with infants with brain injury.17,24,26–29 Brain-derived neurotrophic factor (BDNF), a trophic protein, was associated with severity of injury and worse neurodevelopmental outcomes.24,30,31 Another trophic factor, vascular endothelial growth factor (VEGF), was also associated with worse NE severity, abnormal imaging, and mortality.17,32,33 Research in brain injury biomarkers suggests that a combination of biomarkers is the most effective for providing evidence of injury.9,34,35 However, review of the current literature reports a need for validation studies before these biomarkers can be introduced for routine clinical care.9,10
We investigated a candidate multi-biomarker panel to identify molecules in both plasma and CSF that are associated with NE severity measured by encephalopathy grade, seizures, brain injury by MRI, and neurodevelopmental outcomes at 15–30 months.
Methods:
In this multicenter retrospective cohort study of NE, we identified and analyzed the concentrations of CNS necrosis markers (GFAP, NRGN, Tau), inflammation markers (IL-6, IL-8, IL-10) and trophic-factor markers (BDNF, VEGF) from a cohort of neonates with NE and a cohort of healthy neonatal controls. The study received institutional review board approval at all hospitals, and signed informed consent was obtained from the parent of each participant. Johns Hopkins University Institutional Review Board approved the use of all cohorts in this study.
Control Patient Cohort:
Healthy term neonates (≥36 weeks’ gestation) had plasma samples collected from National Maternity Hospital Dublin and Coombe Women and Infants University Hospital (Trinity College) from 2016–2018 from day of life (DOL) 0–7 with a median collection of DOL 2. The samples were stored in the Trinity Translational Medicine Institute Biobank in Dublin, Ireland. De-identified plasma samples (n=30) from healthy term neonates were analyzed in collaboration with the Neonatal Inflammation and Multiorgan Dysfunction and Brain Injury Research group (NIMBUS) at Trinity College Dublin, Ireland (JHU MTA A33285).
Neonatal Encephalopathy Patient Cohort:
NE was defined as requiring resuscitation at birth and having an abnormal neurological examination. Inclusion criteria were as follows: all infants with NE Sarnat score 2 or 336 requiring TH, NE in the first 48 hours of life without TH, or postnatally diagnosed with brain injury on cranial ultrasound.33,37 Exclusion criteria consisted of maternal substance abuse and major congenital abnormalities. The NE cohort was drawn from National Maternity Hospital Dublin and Children’s National Health System, Washington, D.C. The National Maternity Hospital (Trinity College) NE neonates had plasma and CSF samples collected as previously described. At Children’s National neonates with NE had plasma samples collected from 2012–2016 as part of a prospective study evaluating candidate biomarkers of brain injury in NE. From both studies, a single de-identified plasma sample at enrollment from DOL 0–1 and clinical data (n=155) were analyzed. From the Trinity College NE cohort only, de-identified CSF samples from DOL 0–7, with a median of DOL 3, and clinical data (n=30) were analyzed.
Clinical outcomes of neurologic injury severity were evaluated for significant relationships with candidate biomarker concentrations. To measure clinical severity, degree of encephalopathy was determined by the Sarnat classification and stratified as mild (Sarnat score 0–1) or moderate-to-severe (Sarnat score 2–3).36 Clinical evidence of brain injury was defined as the presence of seizures or severity of injury by MRI on DOL 5–15 according to the Barkovich scale, evaluating the basal ganglia area, watershed area, and the combined basal ganglia/watershed area.38,39 The Barkovich scale was categorized into two groups: score of 0 and scores 1–5.
Neurodevelopmental outcomes were evaluated using cognitive, motor, and language scores of the Bayley Scales of Infant and Toddler Development (Bayley-III) at 15–30 months.40 We evaluated the Bayley-III as both a continuous and binary variable. For our binary analysis, we categorized the scores into two groups normal, scores ≥85 and abnormal, scores <85 or death.41 The patients who died prior to neurodevelopmental follow-up were assigned a score of 39 for the continuous Bayley-III analysis.42
Laboratory Methods:
All primary plasma samples were stored up to 24 hours at 4°C until aliquoted and stored at −80°C. All sample aliquots were exposed to 1–2 freeze/thaws prior to assaying. All assays were performed from 2018–2019 in the same laboratory (Everett Laboratory) at the Johns Hopkins University School of Medicine in Baltimore, MD.
A custom multiplex enzyme-linked immunosorbent assay (ELISA) was developed to measure BDNF, IL-6, IL-8, IL-10, and VEGF simultaneously using robotically spotted capture antibodies on the 96-well plate format (Meso Scale Discovery [MSD], Rockville, MD). Capture antibody-spotted plates were washed with 1xPBS supplemented with 0.05% TWEEN (PBS-T). Calibrators for BDNF, VEGF, and IL-6, IL-8, and IL-10 (MSD) were produced using commercially provided diluent (product number R50AG-2, MSD). The detection antibody cocktail was prepared in commercial diluent (product number R51BA-5, MSD). Plasma and CSF samples were diluted 5-fold. The lower limits of quantification for the BDNF, IL-6, IL-8, IL-10, and VEGF assays were 48.47 pg/mL, 0.47 pg/mL, 0.56 pg/mL, 1.26 pg/mL, and 1.59 pg/mL, respectively, with interassay coefficients of variation of 9.7%, 4.7%, 1.8%, 1.7%, and 2.7%, respectively.
A custom duplex ELISA was developed to measure GFAP and NRGN simultaneously using robotically spotted capture antibodies on the 96-well plate format (MSD, Rockville, MD). The development of the capture antibodies, detection antibodies, and calibrators of NRGN and GFAP have been previously described.21,43 Plasma and CSF samples were diluted 2-fold. The lower limits of quantification for the GFAP and NRGN assays were 0.014 and 0.016 ng/mL, respectively, with interassay coefficients of variation of 2.6% and 2.8%, respectively.
Tau was measured using a commercial ELISA (product number N451LAA-1, MSD). Plasma and CSF samples were diluted 4-fold and assayed according to manufacturer instructions. The lower limits of quantification for Tau was 82.03 pg/mL, with an interassay coefficient of variation of 5.1%.
Statistical Analyses:
Demographic and functional data are presented as median and interquartile range (IQR) or percentages, as appropriate. As demographic, candidate biomarker, and clinical outcome data were not normally distributed for the NE cohort, the Mann Whitney U test was used for categorical variables and Spearman correlation was used for continuous variables. For the binary demographic data, the Fisher exact test was used for comparison. Adjusted analysis was performed by logistic and linear regression using log transformed candidate biomarker concentrations and adjusted for gestational age and sex. For all statistical analyses, a P value of ≤0.05 was considered significant. Statistical analysis was performed using GraphPad Prism (Version 8.0.0 (131); 2018; GraphPad Software, San Diego, CA) and Stata (Version 15, StataCorp LLC, College Station, TX).
Results:
Subject Demographics:
There were 30 healthy term control neonates and 155 neonates with NE available for analysis. From the NE cohort, 155 plasma samples and 30 CSF samples were available for analysis (Table I).
Table I:
Demographic data of healthy term neonates and neonates with neonatal encephalopathy (NE)
| Control Cohort (n=30) | Total NE Cohort (n=155) | Trinity College Cohort (n=57) | Children’s National Cohort (n=98) | p-value | |
|---|---|---|---|---|---|
| Median (IQR) | |||||
| Gestational age (weeks) | 39.1 (37.9, 40.0) | 39.6 (38, 40.7) | 40.7 (40.0, 41.6) | 39.0 (38.0, 40.0) | 0.22a |
| Birthweight (kg) | 3.3 (3.1, 3.6) | 3.3 (2.9, 3.8) | 3.6 (3.2, 4.0) | 3.1 (2.8, 3.6) | 0.73a |
| First blood gas pH | 7.3 (7.2, 7.4) | 7.0 (6.9, 7.1) | 7.0 (6.9, 7.2) | 7.0 (6.8, 7.1) | <0.001a |
| 5-minute Apgar score | 10 (10, 10) | 4 (2, 6) | 5 (3, 7) | 3.5 (2, 5) | <0.001a |
| Sex, n (%) | |||||
| Male | 12 (40) | 88 (57) | 38 (68) | 50 (51) | 0.11b |
| Female | 18 (60) | 66 (43) | 18 (32) | 48 (49) | |
| Mode of Delivery, n (%) | |||||
| Cesarean | 9 (30) | 66 (52) | 20 (36) | 46 (64) | 0.04b |
| Vaginal | 21 (70) | 61 (48) | 35 (64) | 26 (36) | |
| Therapeutic Hypothermia, n (%) | |||||
| Yes | 0 (0) | 135 (88) | 37 (66) | 98 (100) | <0.001b |
| No | 30 (100) | 19 (12) | 19 (34) | 0 (0) | |
NE = neonatal encephalopathy.
Mann Whitney U test was used for comparison of the control cohort and the Total NE cohort.
Fisher’s exact test was used for comparison of control cohort and the Total NE cohort.
Clinical outcomes of the NE cohort are summarized in Table II. The NE cohort were predominately (92%) moderate-to-severe NE (Sarnat score of 2–3). 54% of the NE cohort had seizures. For MRI-derived brain injury, the majority had low Barkovich scores of 0 for basal ganglia (76%), watershed (74%), and combined basal ganglia/watershed area (67%). At 15–30 months, the majority of the NE cohort had normal (Bayley-III score ≥85) Bayley-III cognitive (75%), motor (72%), and language (70%) composite scores. Clinical outcomes were not available for the control cohort.
Table II:
Clinical data of neonates with neonatal encephalopathy (NE)
| Total NE Cohort (n=155) | Trinity College Cohort (n=57) | Children’s National Cohort (n=98) | |
|---|---|---|---|
| Degree of Encephalopathy, n (%) | |||
| Sarnat 0 | 3 (2) | 3 (5) | 0 (0) |
| Sarnat 1 | 9 (6) | 9 (16) | 0 (0) |
| Sarnat 2 | 116 (75) | 36 (63) | 80 (82) |
| Sarnat 3 | 27 (17) | 9 (16) | 18 (18) |
| Seizures, n (%) | |||
| Yes | 74 (54) | 33 (67) | 41 (47) |
| No | 63 (46) | 16 (33) | 47 (53) |
| Brain Injury by MRI: Barkovich Score, n (%) | |||
| Barkovich Basal Ganglia (n=117) | |||
| Barkovich 0 | 89 (76) | 26 (84) | 63 (73) |
| Barkovich 1 | 5 (4) | 0 (0) | 5 (6) |
| Barkovich 2 | 6 (5) | 1 (3) | 5 (6) |
| Barkovich 3 | 9 (8) | 0 (0) | 9 (10) |
| Barkovich 4 | 8 (7) | 4 (13) | 4 (5) |
| Barkovich Watershed (n=117) | |||
| Barkovich 0 | 86 (74) | 22 (71) | 64 (74) |
| Barkovich 1 | 7 (6) | 3 (10) | 4 (5) |
| Barkovich 2 | 7 (6) | 2 (6) | 5 (6) |
| Barkovich 3 | 1 (1) | 0 (0) | 1 (1) |
| Barkovich 4 | 11 (9) | 2 (6) | 9 (10) |
| Barkovich 5 | 5 (4) | 2 (6) | 3 (4) |
| Barkovich Basal Ganglia: Watershed (n=117) | |||
| Barkovich 0 | 78 (67) | 21 (68) | 57 (66) |
| Barkovich 1 | 7 (6) | 1 (3) | 6 (7) |
| Barkovich 2 | 12 (10) | 5 (16) | 7 (8) |
| Barkovich 3 | 16 (14) | 3 (10) | 13 (15) |
| Barkovich 4 | 4 (3) | 1 (3) | 3 (4) |
| Neurodevelopmental Outcomes: Bayley-III, Median (IQR) | |||
| Cognitive Score (n=99) | 99.5 (85, 105) | 105 (95, 110) | 95 (39, 100) |
| Motor Score (n=99) | 97 (76, 107) | 103 (97, 118) | 91.5 (39, 100) |
| Language Score (n=96) | 94 (74, 106) | 100 (86, 112) | 91 (39, 106) |
| Neurodevelopmental Outcomes: Bayley-III, n (%) | |||
| Cognitive Score (n=99) | |||
| Normal (≥85) | 74 (75) | 34 (87) | 40 (67) |
| Abnormal (<85) | 5 (5) | 1 (3) | 4 (7) |
| Dead | 20 (20) | 4 (10) | 16 (27) |
| Motor Score (n=99) | |||
| Normal (≥85) | 71 (72) | 34 (87) | 37 (62) |
| Abnormal (<85) | 8 (8) | 1 (3) | 7 (12) |
| Dead | 20 (20) | 4 (10) | 16 (27) |
| Language Score (n=96) | |||
| Normal (≥85) | 67 (70) | 32 (82) | 35 (61) |
| Abnormal (<85) | 9 (9) | 3 (8) | 6 (11) |
| Dead | 20 (21) | 4 (10) | 16 (28) |
NE = neonatal encephalopathy, Bayley-III = Bayley Scales of Infant and Toddler Development III at 15–30 months. Patients with missing data: Sarnat n=0, Seizures n=18, Barkovich Scores n=38, Bayley-III Cognitive n=56, Motor n=56, Language n=59
Plasma candidate biomarkers in NE and healthy term neonates:
Plasma candidate biomarker concentrations significantly differed in patients with NE compared with controls (Table III; available at www.jpeds.com). CNS necrosis markers NRGN (p=0.03) and Tau (p<0.001) and inflammation markers IL-6 (p<0.001), IL-8 (p<0.001), and IL-10 (p<0.001) were higher, whereas trophic-factors BDNF (p<0.001) and VEGF (p<0.001) were lower in NE compared with controls.
Table III:
Comparison of plasma candidate biomarker concentrations in healthy term neonates and neonates with NE
| Control Cohort Median (IQR) | NE Cohort Median (IQR) | p | |
|---|---|---|---|
| GFAP | 80 (31, 553) | 221 (9.0, 1008) | 0.75 |
| NRGN | 8 (8, 8) | 34 (7, 374) | 0.03 |
| BDNF | 1376.9 (867.8, 2629.0) | 407.3 (152.5, 1161.0) | <0.001 |
| IL-6 | 4.9 (2.6, 14.3) | 28.4 (9.8, 119.9) | <0.001 |
| IL-8 | 32.2 (24.4, 47.0) | 113.5 (52.3, 394.9) | <0.001 |
| IL-10 | 0.7 (0.3, 2.2) | 8.5 (2.2, 48.8) | <0.001 |
| VEGF | 276.1 (142.3, 319.9) | 12.9 (0.9, 60.7) | <0.001 |
| Tau | 111.4 (32.8, 182.3) | 243.1 (115.1, 541.0) | <0.001 |
NE = neonatal encephalopathy. All analyses used the Mann-Whitney U test for comparisons of non-parametric data. Candidate biomarker units are pg/mL. Duplex (GFAP and NRGN): Control n=24, NE n=141; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): Control n=29, NE n=141; Tau: Control n=27, NE n=123
Candidate biomarkers and clinical outcomes of brain injury:
In univariate analysis of plasma candidate biomarkers, higher Tau (p<0.001), GFAP (p=0.05), and NRGN (p=0.03) and lower BDNF (p=0.002) and VEGF (p=0.05) were associated with moderate-to-severe encephalopathy (Table IV; available at www.jpeds.com). IL-6, IL-8, and IL-10 were all higher but not significantly. When adjusted for gestational age and sex, Tau (p=0.002), GFAP (p=0.03), and NRGN (p=0.05) were directly related and BDNF (p=0.04) was inversely related to severity of encephalopathy (Table V). There were no significant associations of CSF candidate biomarkers with moderate-to-severe encephalopathy in univariate or adjusted analysis (Table VI; available at www.jpeds.com).
Table IV:
Univariate analysis of plasma candidate biomarker concentrations of neonates with NE and clinical encephalopathy
| Mild (Sarnat 0–1) Median (IQR) | Moderate-to-Severe (Sarnat 2–3) Median (IQR) |
p | |
|---|---|---|---|
| GFAP | 9 (9, 45) | 251 (9, 1154) | 0.05 |
| NRGN | 7 (7, 7) | 40 (7, 403) | 0.03 |
| BDNF | 1596.5 (1081.2, 2324.1) | 387.2 (136.1, 1006.5) | 0.002 |
| IL-6 | 21.4 (7.2, 39.1) | 29.7 (9.8, 148.5) | 0.57 |
| IL-8 | 64.7 (46.1, 104.4) | 119.6 (52.6, 404.4) | 0.11 |
| IL-10 | 4.6 (1.3, 16.7) | 9.2 (2.3, 59.5) | 0.23 |
| VEGF | 87.0 (0.9, 330.2) | 11.7 (0.9, 55.3) | 0.05 |
| Tau | 45.8 (45.8, 103.6) | 276.7 (135.5, 554.4) | <0.001 |
NE = neonatal encephalopathy. Univariate analyses used the Mann-Whitney U test for comparisons of non-parametric data. Candidate biomarker units are pg/mL. Duplex (GFAP and NRGN): Mild n=11, Moderate-to-Severe n=130; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): Mild n=10, Moderate-to-Severe n=131; Tau: Mild n=11, Moderate-to-Severe n=112.
Table V:
Association of plasma candidate biomarker concentrations of neonates with NE and clinical encephalopathy, seizure occurrence, and brain injury by MRI*
| Clinical Encephalopathy | Seizure Occurrence | MRI Barkovich Score | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Basal Ganglia | Watershed | BG:WS Score | ||||||||
| Coefficient (95% CI) | p | Coefficient (95% CI) | p | Coefficient (95% CI) | p | Coefficient (95% CI) | p | Coefficient (95% CI) | p | |
| GFAP | 0.38 (0.04–0.71) | 0.03 | −0.007 (−0.15–0.14) | 0.93 | 0.11 (−0.07–0.30) | 0.22 | 0.08 (−0.10–0.26) | 0.39 | 0.003 (−0.16–0.17) | 0.98 |
| NRGN | 0.48 (−0.001–0.95) | 0.05 | −0.06 (−0.22–0.11) | 0.50 | 0.14 (−0.06–0.34) | 0.16 | 0.06 (−0.13–0.25) | 0.54 | 0.04 (−0.14–0.22) | 0.70 |
| BDNF | −0.61 (−1.18–−0.03) | 0.04 | −0.09 (−0.29–0.11) | 0.37 | −0.23 (−0.47–0.01) | 0.06 | −0.20 (−0.45–0.04) | 0.10 | −0.17 (−0.40–0.06) | 0.14 |
| IL-6 | 0.04 (−0.24–0.31) | 0.80 | −0.19 (−0.36–0.03) | 0.02 | 0.03 (−0.18–0.23) | 0.81 | 0.03 (−0.18–0.24) | 0.76 | −0.004 (−0.19–0.18) | 0.97 |
| IL-8 | 0.12 (−0.18–0.43) | 0.43 | −0.002 (−0.19–0.19) | 0.98 | 0.09 (−0.18–0.35) | 0.51 | 0.27 (−0.04–0.58) | 0.09 | 0.11 (−0.13–0.36) | 0.37 |
| IL-10 | 0.16 (−0.16–0.49) | 0.32 | −0.008 (−0.18–0.16) | 0.93 | −0.05 (−0.27–0.17) | 0.69 | 0.11 (−0.12–0.33) | 0.36 | −0.06 (−0.26–0.14) | 0.56 |
| VEGF | −0.23 (−0.53–0.07) | 0.14 | −0.02 (−0.16–0.13) | 0.84 | −0.06 (−0.23–0.11) | 0.50 | −0.09 (−0.26–0.08) | 0.30 | −0.04 (−0.20–0.12) | 0.65 |
| Tau | 1.46 (0.53–2.38) | 0.002 | 0.08 (−0.23–0.39) | 0.62 | 0.74 (0.26–1.22) | 0.002 | 0.47 (0.05–0.88) | 0.03 | 0.60 (0.18–1.03) | 0.006 |
NE = neonatal encephalopathy.
Adjusted for gestational age and sex, the natural log of candidate biomarker concentrations were used in adjusted analyses. Clinical Encephalopathy: Duplex (GFAP and NRGN): n=141; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): n=141; Tau: n=123. Seizure Occurrence: Duplex (GFAP and NRGN): n=125; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): n=125; Tau: n=109. MRI: Duplex (GFAP and NRGN): n=107; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): n=107; Tau: n=93
Table VI:
Association of CSF candidate biomarker concentrations of neonates with NE and clinical encephalopathy
| Univariate Analysis | Adjusted Logistic Regression* | ||||
|---|---|---|---|---|---|
| Mild (Sarnat 0–1) Median (IQR) | Moderate-to-Severe (Sarnat 2–3) Median (IQR) | p | Coefficient (95% CI) | p | |
| GFAP | 322 (220, 395) | 247 (156, 415) | 0.54 | −0.118 (−0.953–0.716) | 0.78 |
| NRGN | 7 (7, 24) | 7 (7, 7) | 0.50 | −0.783 (−2.439–0.873) | 0.35 |
| BDNF | 30.3 (30.3, 30.3) | 30.3 (30.3, 30.3) | 0.57 | - | |
| IL-6 | 4.0 (3.1, 142.6) | 3.8 (0.2, 36.2) | 0.46 | −0.135 (−0.580–0.311) | 0.55 |
| IL-8 | 369.9 (160.4, 501.2) | 203.5 (78.4, 606.1) | 0.76 | −0.122 (−0.777–0.532) | 0.71 |
| IL-10 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) | 0.70 | - | |
| VEGF | 4.5 (2.0, 6.6) | 4.9 (3.3, 9.6) | 0.46 | 0.603 (−0.784–1.990) | 0.39 |
| Tau | 3223.9 (56.6, 2.8e+04) | 2626.1 (56.6, 6721.5) | 0.80 | −0.016 (−0.446–0.415) | 0.94 |
NE = neonatal encephalopathy, CSF = cerebrospinal fluid. Univariate analyses used the Mann-Whitney U test for comparisons of non-parametric data. Candidate biomarker units are pg/mL. Duplex (GFAP and NRGN): Mild n=4, Moderate-to-Severe n=26; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): Mild n=4, Moderate-to-Severe n=26; Tau: Mild n=4, Moderate-to-Severe n=23.
Adjusted for gestational age and sex, the natural log of candidate biomarker concentrations were used in adjusted analyses.
In univariate analysis of plasma candidate biomarkers and seizure occurrence, lower IL-6 (p=0.02) was associated with seizures (Table VII; available at www.jpeds.com). In adjusted analysis, IL-6 showed an inverse relationship (p=0.02) with seizure occurrence (Table V). There were no significant associations of CSF candidate biomarkers with seizures in univariate or adjusted analysis (Table VIII; available at www.jpeds.com).
Table VII:
Univariate Analysis of plasma candidate biomarker concentrations of neonates with NE and seizure occurrence
| No Median (IQR) | Yes Median (IQR) | p | |
|---|---|---|---|
| GFAP | 224 (9, 878) | 153 (9, 1008) | 0.96 |
| NRGN | 57 (7, 549) | 17 (7, 313) | 0.64 |
| BDNF | 373.2 (149.6, 1112.5) | 480.5 (183.2, 1525.5) | 0.52 |
| IL-6 | 43.7 (13.4, 295.5) | 22.7 (6.5, 66.8) | 0.02 |
| IL-8 | 100.8 (53.6, 341.6) | 130.7 (47.5, 439.3) | 0.75 |
| IL-10 | 11.4 (2.9, 36.4) | 7.6 (1.3, 59.5) | 0.39 |
| VEGF | 10.0 (0.2, 76.5) | 14.3 (0.9, 52.6) | 0.79 |
| Tau | 251.9 (123.8, 528.2) | 239.8 (119.2, 554.5) | 0.90 |
NE = neonatal encephalopathy. Univariate analyses used the Mann-Whitney U test for comparisons of non-parametric data. Candidate biomarker units are pg/mL. Duplex (GFAP and NRGN): No n=58, Yes n=67; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): No n=60, Yes n=65; Tau: No n=51, Yes n=58.
Table VIII:
Association of CSF candidate biomarker concentrations of neonates with NE and seizure occurrence
| Univariate Analysis | Adjusted Logistic Regression* | ||||
|---|---|---|---|---|---|
| No Median (IQR) | Yes Median (IQR) | p-value | Coefficient (95% CI) | p-value | |
| GFAP | 258 (251, 345) | 202 (156, 636) | 0.57 | −0.048 (−0.682–0.587) | 0.88 |
| NRGN | 7 (7, 7) | 7 (7, 7) | 0.68 | 0.273 (−1.157–1.702) | 0.71 |
| BDNF | 30.3 (30.3, 30.3) | 30.3 (30.3, 30.3) | 0.48 | - | |
| IL-6 | 3.1 (0.7, 196.2) | 4.5 (0.2, 36.2) | 0.68 | −0.059 (−0.387–0.270) | 0.73 |
| IL-8 | 300.5 (81.3, 503.9) | 203.5 (74.2, 606.1) | 0.78 | 0.036 (−0.362–0.433) | 0.86 |
| IL-10 | 1.0 (1.0, 1.0) | 1.0 (1.0, 1.0) | 0.48 | - | |
| VEGF | 5.5 (2.8, 7.3) | 3.8 (2.6, 6.9) | 0.76 | −0.008 (−0.770–0.754) | 0.98 |
| Tau | 2337.5 (56.6, 6391.4) | 5290.6 (56.6, 7552.7) | 0.46 | 0.127 (−0.217–0.471) | 0.47 |
NE = neonatal encephalopathy, CSF = cerebrospinal fluid. Univariate analyses used the Mann-Whitney U test for comparisons of non-parametric data. Candidate biomarker units are pg/mL. Duplex (GFAP and NRGN): No n=9, Yes n=18; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): No n=9, Yes n=18; Tau: No n=9, Yes n=15.
Adjusted for gestational age and sex, the natural log of candidate biomarker concentrations were used in adjusted analyses.
Candidate biomarkers and brain injury by MRI:
In adjusted analysis, only plasma Tau was directly related to Barkovich basal ganglia (p=0.002), watershed (p=0.03), and basal ganglia/watershed (p=0.006) scores (Table V). CSF candidate biomarkers did not have any significant relationships with MRI abnormalities (Table IX; available at www.jpeds.com).
Table IX:
Association of CSF candidate biomarker concentrations of neonates with NE and brain injury by MRI
| Basal Ganglia | Watershed | BG:WS Score | ||||
|---|---|---|---|---|---|---|
| Coefficient (95% CI) | p | Coefficient (95% CI) | p | Coefficient (95% CI) | p | |
| GFAP | 0.53 (−0.58–1.63) | 0.35 | 0.10 (−0.75–0.96) | 0.82 | 1.46 (−0.43–3.346) | 0.13 |
| NRGN | - | 0.37 (−1.15–1.90) | 0.63 | - | ||
| BDNF | - | - | - | |||
| IL6 | 0.63 (−0.10–1.35) | 0.09 | 0.04 (−0.34–0.41) | 0.84 | 0.10 (−0.30–0.49) | 0.63 |
| IL8 | 0.53 (−0.28–1.35) | 0.20 | −0.02 (−0.45–0.42) | 0.95 | 0.07 (−0.43–0.57) | 0.78 |
| IL10 | - | - | - | |||
| VEGF | −0.07 (−1.07–0.93) | 0.89 | 0.00 (−0.79–0.79) | 0.99 | −0.03 (−0.82–0.76) | 0.95 |
| Tau | 0.77 (−0.53–2.07) | 0.25 | 0.11 (−0.34–0.55) | 0.64 | 0.14 (−0.33–0.61) | 0.55 |
NE = neonatal encephalopathy. All analyses used adjusted logistic regression, adjusted for gestational age and sex, the natural log of biomarker concentrations were used in adjusted analyses. Candidate biomarker units are pg/mL. Duplex (GFAP and NRGN): n=20; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): n=20; Tau: n=18
Candidate biomarkers and neurodevelopmental outcomes:
In univariate analysis of plasma candidate biomarkers, Tau and IL-6 negatively correlated with Bayley-III cognitive (p=0.02, p=0.05) and motor (p=0.02, p=0.03) composite scores (Table X; available at www.jpeds.com). IL-8 negatively correlated with cognitive (p=0.005), motor (p<0.001), and language (p=0.02) composite scores. IL-10 negatively correlated with motor composite scores only (p=0.03). VEGF positively correlated with motor composite scores (p=0.03). In univariate analysis of CSF candidate biomarkers, NRGN and IL-6 negatively correlated with Bayley-III cognitive (p=0.04, p=0.05) and motor (p=0.03, p=0.02) scores, respectively (Table XI; available at www.jpeds.com). IL-8 negatively correlated with motor scores only (p=0.03).
Table X:
Univariate analysis of plasma candidate biomarker concentrations of neonates with NE and Bayley Scales of Infant and Toddler Development III (Bayley-III) scores at 15–30 months
| Cognitive | Motor | Language | ||||
|---|---|---|---|---|---|---|
| rho | p | rho | p | rho | p | |
| GFAP | −0.04 | 0.72 | −0.09 | 0.38 | 0.01 | 0.95 |
| NRGN | −0.10 | 0.36 | −0.16 | 0.12 | −0.05 | 0.67 |
| BDNF | 0.16 | 0.14 | 0.13 | 0.23 | 0.03 | 0.76 |
| IL-6 | −0.21 | 0.05 | −0.23 | 0.03 | −0.15 | 0.17 |
| IL-8 | −0.29 | 0.005 | −0.40 | <0.001 | −0.25 | 0.02 |
| IL-10 | −0.20 | 0.06 | −0.23 | 0.03 | −0.17 | 0.11 |
| VEGF | 0.16 | 0.13 | 0.23 | 0.03 | 0.14 | 0.20 |
| Tau | −0.27 | 0.02 | −0.26 | 0.02 | −0.16 | 0.17 |
NE = neonatal encephalopathy. Univariate analyses used Spearman correlation for non-parametric data. Candidate biomarker units are pg/mL. Duplex (GFAP and NRGN): Cognitive and Motor n=93, Language n=90; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): Cognitive and Motor n=92, Language n=89; Tau: Cognitive and Motor n=81, Language n=80.
Table XI:
Univariate Analysis of CSF candidate biomarker concentrations of neonates with NE and Bayley Scales of Infant and Toddler Development III (Bayley-III) scores at 15–30 months
| Cognitive | Motor | Language | ||||
|---|---|---|---|---|---|---|
| rho | p | rho | p | rho | p | |
| GFAP | −0.31 | 0.19 | −0.27 | 0.25 | −0.12 | 0.62 |
| NRGN | −0.46 | 0.04 | −0.48 | 0.03 | −0.39 | 0.09 |
| BDNF | 0.12 | 0.61 | 0.10 | 0.69 | 0.03 | 0.90 |
| IL-6 | −0.44 | 0.05 | −0.51 | 0.02 | −0.26 | 0.26 |
| IL-8 | −0.40 | 0.08 | −0.50 | 0.03 | −0.25 | 0.28 |
| IL-10 | −0.34 | 0.14 | −0.34 | 0.14 | −0.34 | 0.14 |
| VEGF | −0.06 | 0.80 | −0.06 | 0.79 | −0.07 | 0.78 |
| Tau | −0.23 | 0.38 | −0.25 | 0.33 | −0.07 | 0.80 |
NE = neonatal encephalopathy. Univariate analyses used Spearman correlation for non-parametric data. Candidate biomarker units are pg/mL. Duplex (GFAP and NRGN): Cognitive and Motor n=93, Language n=90; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): Cognitive and Motor n=92, Language n=89; Tau: Cognitive and Motor n=81, Language n=80.
In adjusted analysis of plasma candidate biomarkers, only Tau was associated with decreased cognitive (p=0.03) and motor (p=0.03) outcomes (Table XII). In adjusted analysis of CSF candidate biomarkers, NRGN was inversely related to all three outcomes of cognitive (p=0.001), motor (p<0.001), and language (p=0.007) measures (Table XII). GFAP, IL-6, and IL-10 were inversely related to cognitive (p=0.02, p=0.02, p=0.02) and motor (p=0.04, p=0.009, p=0.008) outcomes, respectively. IL-8 was inversely related to motor outcomes only (p=0.03).
Table XII:
Association of plasma and CSF candidate biomarker concentrations of neonates with NE and Bayley Scales of Infant and Toddler Development III (Bayley-III) scores at 15–30 months*
| Plasma | CSF | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Cognitive | Motor | Language | Cognitive | Motor | Language | |||||||
| Coefficient (95% CI) | p | Coefficient (95% CI) | p | Coefficient (95% CI) | p | Coefficient (95% CI) | p | Coefficient (95% CI) | p | Coefficient (95% CI) | p | |
| GFAP | −0.03 (−2.8–2.7) | 0.98 | −0.4 (−3.2–2.5) | 0.80 | 0.3 (−2.8–3.3) | 0.87 | −9.2 (−17−-1.9) | 0.02 | −8.7 (−17−-0.5) | 0.04 | −7.0 (−15–1.1) | 0.09 |
| NRGN | −0.07 (−3.4–3.3) | 0.97 | −0.6 (−4.1–2.9) | 0.73 | 0.2 (−3.5–3.9) | 0.90 | −22.4 (−34−-11.3) | 0.001 | −25.3 (−37−-13.8) | <0.001 | −19.1 (−32−-6.0) | 0.007 |
| BDNF | 1.4 (−2.3–5.2) | 0.45 | 1.6 (−2.3–5.4) | 0.43 | 0.9 (−3.3–5.0) | 0.68 | 4.2 (−9.5–18.0) | 0.52 | 5.0 (−9.8–19.6) | 0.49 | 1.9 (−12.4–16.2) | 0.78 |
| IL-6 | −2.2 (−5.2–0.8) | 0.16 | −2.2 (−5.3–0.9) | 0.17 | −2.5 (−5.8–0.8) | 0.14 | −4.4 (−8.0−-1.0) | 0.02 | −5.1 (−8.7- −1.5) | 0.009 | −3.6 (−7.5–0.3) | 0.07 |
| IL-8 | −2.6 (−6.2–1.0) | 0.16 | −3.0 (−6.7–0.9) | 0.13 | −3.2 (−7.2–0.7) | 0.11 | −2.9 (−7.1–1.3) | 0.16 | −4.8 (−8.9- −0.6) | 0.03 | −2.5 (−7.0–1.9) | 0.25 |
| IL-10 | −0.3 (−4.0–3.5) | 0.89 | −0.6 (−4.5–3.3) | 0.76 | −1.6 (−5.6–2.5) | 0.45 | −34.6 (−63−-5.9) | 0.02 | −42.0 (−71−-12.7) | 0.008 | −20.2 (−54–13.0) | 0.22 |
| VEGF | 0.5 (−2.5–3.4) | 0.75 | 0.9 (−2.2–3.9) | 0.57 | 1.0 (−2.3–4.2) | 0.56 | −0.4 (−9.6–8.8) | 0.93 | 0.08 (−9.8–10.0) | 0.99 | −1.5 (−11.0–8.1) | 0.75 |
| Tau | −5.6 (−10.6−-0.5) | 0.03 | −6.1 (−11.5−-0.7) | 0.03 | −5.4 (−10.9–0.2) | 0.06 | −2.1 (−6.8–2.7) | 0.37 | −2.6 (−7.5–2.3) | 0.27 | −0.8 (−5.8–4.1) | 0.73 |
NE = neonatal encephalopathy.
Adjusted for gestational age and sex, the natural log of candidate biomarker concentrations were used in adjusted analyses. Plasma: Duplex (GFAP and NRGN): n=93; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): n=92; Tau: n=81. CSF: Duplex (GFAP and NRGN): n=20; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): n=20; Tau: n=17.
In univariate analysis of plasma candidate biomarkers and binary neurodevelopmental outcomes (abnormal [Bayley-III <85 or death] or normal [≥85]), higher Tau and IL-8 were associated with abnormal cognitive (p=0.03, p=0.05), motor (p=0.01, p=0.009), and language (p=0.04, p=0.005) outcomes, respectively (Table XIII; available at www.jpeds.com). In adjusted analysis, greater than 1 standard deviation of increased Tau between individuals presented a 1.76 times increased likelihood for abnormal cognitive outcomes (p=0.04), 1.82 times increased likelihood for abnormal motor outcomes (p=0.03), and 1.71 times increased likelihood for abnormal language outcomes (p=0.05).
Table XIII:
Association of plasma candidate biomarker concentrations of neonates with NE and binary outcomes of normal and abnormal Bayley Scales of Infant and Toddler Development III (Bayley-III) scores at 15–30 months
| Bayley-III Score | Univariate Analysis | Adjusted Logistic Regression* | |||
|---|---|---|---|---|---|
| Normal (≥85) Median (IQR) | Abnormal (<85 or death) Median (IQR) | p-value | Odds Ratio of Standard Deviation (95% CI) | p-value | |
| Cognitive | |||||
| GFAP | 300.4 (8.7, 1154.2) | 284.6 (8.7, 821.8) | 0.68 | 0.88 (0.55–1.42) | 0.59 |
| NRGN | 35.9 (7.1, 498.7) | 49.9 (16.5, 243.2) | 0.68 | 0.94 (0.58–1.53 ) | 0.81 |
| BDNF | 444.6 (134.6, 1802.3) | 317.3 (77.0, 692.5) | 0.18 | 0.75 (0.46–1.22) | 0.25 |
| IL-6 | 28.8 (11.5, 116.8) | 35.1 (9.5, 2425.8) | 0.31 | 1.32 (0.80–2.16) | 0.28 |
| IL-8 | 115.3 (59.6, 462.6) | 324.8 (94.5, 2458.4) | 0.05 | 1.26 (0.73–2.15) | 0.41 |
| IL-10 | 11.6 (1.7, 60.5) | 13.6 (1.9, 111.8) | 0.76 | 0.92 (0.55–1.54) | 0.76 |
| VEGF | 11.7 (0.9, 82.2) | 9.3 (0.5, 43.8) | 0.35 | 0.89 (0.54–1.49) | 0.66 |
| Tau | 218.0 (102.4, 541.6) | 454.1 (210.8, 1410.2) | 0.03 | 1.76 (1.02–3.06) | 0.04 |
| Motor | |||||
| GFAP | 281.9 (8.7, 1154.2) | 429.1 (8.7, 1039.1) | 0.87 | 0.91 (0.57–1.45) | 0.69 |
| NRGN | 35.1 (7.1, 498.7) | 59.4 (16.5, 314.2) | 0.66 | 0.95 (0.59–1.52) | 0.82 |
| BDNF | 434.6 (134.6, 1720.7) | 414.6 (77.0, 868.7) | 0.37 | 0.80 (0.50–1.28) | 0.35 |
| IL-6 | 28.8 (10.9, 116.8) | 33.4 (10.9, 2365.0) | 0.26 | 1.34 (0.83–2.14) | 0.23 |
| IL-8 | 108.9 (53.9, 398.7) | 360.6 (106.8, 2458.4) | 0.009 | 1.43 (0.83–2.48) | 0.20 |
| IL-10 | 10.9 (1.4, 37.0) | 18.8 (2.8, 134.4) | 0.34 | 1.08 (0.66–1.75) | 0.77 |
| VEGF | 12.6 (0.9, 83.6) | 10.5 (0.5, 43.8) | 0.32 | 0.91 (0.56–1.48) | 0.70 |
| Tau | 215.0 (95.4, 541.0) | 412.8 (210.8, 1410.2) | 0.01 | 1.82 (1.06–3.12) | 0.03 |
| Language | |||||
| GFAP | 300.4 (8.7, 1154.2) | 99.1 (8.7, 902.6) | 0.46 | 0.79 (0.49–1.26) | 0.32 |
| NRGN | 34.3 (7.11, 498.7) | 38.0 (16.5, 208.2) | 0.98 | 0.83 (0.51–1.35) | 0.45 |
| BDNF | 446.6 (136.1, 1808.5) | 372.5 (77.0, 724.9) | 0.19 | 0.76 (0.47–1.22) | 0.25 |
| IL-6 | 27.4 (10.8, 113.7) | 35.1 (10.9, 2365.0) | 0.21 | 1.39 (0.86–2.26) | 0.18 |
| IL-8 | 101.3 (49.9, 390.7) | 360.6 (114.4, 1836.6) | 0.005 | 1.51 (0.87–2.63) | 0.15 |
| IL-10 | 9.2 (1.3, 31.6) | 22.1 (2.8 134.4) | 0.19 | 1.23 (0.75–2.01) | 0.42 |
| VEGF | 14.8 (0.9, 83.6) | 7.2 (0.2, 43.8) | 0.18 | 0.83 (0.51–1.36) | 0.46 |
| Tau | 218.0 (102.4, 541.0) | 412.8 (209.1, 1410.2) | 0.04 | 1.71 (1.00–2.93) | 0.05 |
NE = neonatal encephalopathy. Univariate analyses used the Mann-Whitney U test for comparisons of non-parametric data. Candidate biomarker units are pg/mL. Duplex (GFAP and NRGN): Cognitive: Normal n=69, Abnormal n=24, Motor: Normal n=66, Abnormal n=27, Language: Normal n=63, Abnormal n=27; Multiplex (BDNF, IL-6, IL-8, IL-10, VEGF): Cognitive: Normal n=68, Abnormal n=24, Motor: Normal n=64, Abnormal n=28, Language: Normal n=61, Abnormal n=28; Tau: Cognitive: Normal n=57, Abnormal n=19, Motor: Normal n=59, Abnormal n=22, Language: Normal n=55, Abnormal n=22
Adjusted for gestational age and sex, the natural log of candidate biomarker concentrations were used in adjusted analyses.
Discussion:
In this multicenter retrospective cohort study of NE, we demonstrated a candidate multi-biomarker panel of CNS necrosis, inflammatory, and trophic-factor proteins differentiated neonates with NE from healthy term neonates within the first 24 hours of life. Moreover, this candidate biomarker panel, with the best performance from Tau, differentiated severity of NE measured by clinical encephalopathy, seizures, brain injury by MRI, and neurodevelopmental outcomes at 15–30 months. Although previous research focused on candidate biomarkers such as cytokines, Tau, and GFAP, our study contributes novel investigation of less studied candidate biomarkers, NRGN, VEGF, and BDNF, providing important performance comparisons.
We found that CNS necrosis markers (NRGN and Tau) and inflammation markers (IL-6, IL-8, and IL-10) were higher, whereas trophic factors (BDNF and VEGF) were lower in NE compared with controls. Higher IL-627–29 and Tau,23 and lower VEGF44 in neonates with NE compared with controls is consistent with previous smaller, single center studies. The findings of higher NRGN21,22 and lower BDNF45–47 in patients with brain injury compared with healthy controls has also been reported in adults with traumatic brain injury and delirium. Conversely, others showed higher BDNF in neonates with asphyxia compared with controls using cord blood samples.30,31 Olin et al found that cord blood is not representative of neonatal blood, and therefore our use of plasma samples from the first 24 hours of life is a strength of our study.48 Additionally, timing variation of sample collection and TH may have also contributed to discrepancies, as cooling may affect candidate biomarker concentrations.32,44 However, because the standard of care is to begin TH within 6 hours of birth, and the plasma samples in our study span DOL 0–1, TH was considered a mediator, not a confounder, and was not included in the adjusted analysis. The median plasma concentrations of GFAP and Tau are higher in the control cohort than the infants with mild encephalopathy in the NE cohort. This difference was driven by three control outliers not excluded from analysis that each had elevated GFAP, Tau and neurogranin. In addition, although the control cohort had no overt clinical brain injury at birth, they were neonates admitted to the NICU and could have had subclinical brain injury to account for the elevated GFAP and Tau. Finally, there are multiple factors as previously discussed including the influence of TH and timing variation of sample collection that may contribute to this observation.
The identification of early candidate biomarkers to discriminate NE severity is becoming increasingly important with potential adjuvant therapies to TH, such as xenon, erythropoietin, and stem cells.24,49–52 In our study, using adjusted analysis, plasma Tau, GFAP, NRGN, and BDNF differentiated between mild and moderate-to-severe encephalopathy. The higher Tau, GFAP, and NRGN concentrations in moderate-to-severe encephalopathy supported the hypothesis that these CNS necrosis markers are indicators of acute and possibly ongoing neuronal injury.11,17,23,35 Meanwhile, BDNF was lower in moderate-to-severe encephalopathy. BDNF is a neuronal survival factor, suggesting an inadequate concentration to protect the brain in more severe injuries.53–55 CSF candidate biomarkers did not differentiate encephalopathy severity in our study, although others reported significant results for CSF GFAP,56,57 CSF IL-6,29 and CSF VEGF.32 This could possibly be due to availability of CSF samples at a single time point compared with 0–7 days of life.
Seizures are a common sequelae of NE and infants with seizures are associated with worse neurodevelopmental outcomes or death.7,8,36,58 Seizures were common in our cohort, and neonates with seizures had predominately moderate-to-severe NE (99% Sarnat score 2–3). We found lower plasma IL-6 concentrations could differentiate and were predictive of seizures in neonates with NE. Conversely, Numis et al showed higher IL-6 was associated with epilepsy in newborns with NE.59 This discrepancy could reflect variation in the dynamic inflammatory process and possible dysregulation in severely injured neonates. Furthermore, our study did not control for infectious etiologies, due to limited sample size and availability of clinical data. Further evaluation of inflammatory cytokines considering infectious etiologies is warranted in larger cohorts. Candidate biomarkers in CSF were not able to differentiate seizure occurrence.
Evidence of structural injury, especially the basal ganglia, in NE is well described.38 We found with adjusted analysis that plasma Tau directly related to Barkovich score for basal ganglia, watershed, and combined basal ganglia/watershed area injury. Previous research supports our findings that plasma Tau is associated with worse brain injury severity on MRI.24 Tau is an axonal protein and particularly enriched in the brain white matter.60,61 The presence of elevated Tau suggests additional white matter injury in NE, which has been suspected based on clinical outcomes and MRI.62 However, MRI is not diagnostic until 24 hours after the injury and for neonates treated with TH, TH devices are not MRI-compatible delaying MRI use.10,63,64 Our data suggest that plasma Tau, collected on DOL 0–1, could be an effective predictor of brain injury and help to fill this crucial time gap for intervention. CSF candidate biomarkers were not associated with brain injury measured on MRI.
When evaluating neurodevelopmental outcomes related to brain injury severity, we found with adjusted analysis, that plasma Tau and CSF GFAP, NRGN, IL-6, IL-8, and IL-10 were negatively associated with Bayley-III scores at 15–30 months. Specifically, higher plasma Tau collected on DOL 0–1 predicted abnormal cognitive and motor outcomes at 15–30 months. Adding multiple candidate biomarkers to the regression model did not significantly improve the association of outcomes beyond that seen with Tau. Other studies using plasma and CSF also demonstrated higher levels of GFAP, IL-6, IL-8, and Tau were associated with abnormal neurodevelopmental outcomes at 15–30 months.14,15,17,24,25,29,65,66 Previous research also supports that lower VEGF is associated with abnormal neurodevelopmental outcomes.32 Our study suggests that higher levels of CNS necrosis, especially Tau, and inflammatory markers could predict poor neurodevelopmental performance in the future.
Furthermore, we analyzed the baseline characteristics of infants without reported neurodevelopmental assessments, to evaluate the impact of candidate biomarker association with abnormal outcomes because the data from neonates without MRI Barkovich scores (n=38) and without Bayley-III scores at 15–30 months (n=56) were not included in analysis. Those without reported MRI Barkovich scores appeared to be healthier infants with greater gestational age, and neonates with higher chance of vaginal delivery, lower rates of TH, and lower proportion of moderate-to-severe Sarnat Scores. Those without reported Bayley-III scores importantly did not differ in degree of encephalopathy, and therefore should not affect our conclusions about abnormal neurodevelopmental outcomes. Future studies should include analyses of factors that can impact follow-up rates and neurodevelopmental outcomes at follow-up, including socioeconomic status.
Limitations of this study include small sample size, heterogeneity of NE cases, a single time point for some samples, sample variation over 24 hours, and unavailability of clinical data, which limited the ability to evaluate some outcomes. The NE criteria for this study included infants with NE identified <6 hours with TH, <48 hours without TH, and postnatally with brain injury on cranial ultrasound. This reflects a heterogeneity of cases; however, all infants had perinatal asphyxia. Due to the limited availability of samples and clinical data, the absence of CSF samples from healthy term neonates precluded any comparison of CSF candidate biomarkers in the NE cohort compared with controls. Additionally, CSF candidate biomarker analysis of neurodevelopmental outcomes was limited by a lack of CSF samples from neonates with abnormal (<85) Bayley-III scores (n=1). In addition for the CSF analysis, there was a small number of neonates with mild encephalopathy (n=4) compared with moderate-to-severe encephalopathy (n=26). Lastly, the inclusion of CSF samples spanning DOL 0–7 (median DOL 3) may influence the associations with clinical outcomes.
In conclusion, the ideal biomarker for identifying, stratifying, and monitoring NE would need to be stable, measurable at a high sensitivity in an easy-to-access biofluid, have peak concentrations early in life, and have the ability to discriminate NE severity and predict neurodevelopmental outcomes. Our study provided novel insight into a selection of candidate biomarkers that fit these optimal criteria for NE, with Tau as the best performer in multiple measures of brain injury and outcomes. Furthermore, plasma candidate biomarkers were able to identify neonates with NE, discriminate clinical severity, and predict seizures, brain injury by MRI, and neurodevelopmental outcomes at 15–30 months. Our study also identified potential adjunctive therapies for NE. Larger validation studies are needed to further investigate this candidate biomarker panel and its implementations into clinical settings.
Acknowledgements
We thank the patients and their families for their participation and contributions to this study. We thank the Everett research group at Johns Hopkins University School of Medicine for their support and contributions. We thank the Johns Hopkins University School of Medicine Scholarly Concentration mentor Dr Meredith Atkinson and the Johns Hopkins University School of Medicine Dean’s Funding for their support and contributions.
Supported by NIH NICHD (R01HD086058 [to A.E. and F.N.]); Health Research Board, Ireland; Trinity College Dublin; National Children’s Research Centre; Clinical and Translational Science Institute at Children’s National (UL1TR000075, 1KL2RR031987-01 [to A.M.]); and the Intellectual and Developmental Disabilities Research Consortium (NIH P30HD040677 [to A.M.]). Under a license agreement between ImmunArray Ltd. and the Johns Hopkins University, the University and A.E. are entitled to royalties on an invention described in this study and discussed in this publication. This arrangement has been reviewed and approved by the Johns Hopkins University in accordance with its conflict of interest policies. The other authors declare no conflicts of interest.
Abbreviations
- BDNF
brain-derived neurotrophic factor
- Bayley-III
Bayley Scales of Infant and Toddler Development III
- CNS
central nervous system
- CSF
cerebrospinal fluid
- DOL
day of life
- ELISA
enzyme-linked immunosorbent assay
- GFAP
glial fibrillary acidic protein
- HIE
hypoxic-ischemic encephalopathy
- IL-6
interleukin-6
- IL-8
interleukin-8
- IL-10
interleukin-10
- IQR
interquartile range
- MRI
magnetic resonance imaging
- NRGN
neurogranin
- VEGF
vascular endothelial growth factor
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Molloy EJ, Bearer C. Neonatal encephalopathy versus Hypoxic-Ischemic Encephalopathy. Pediatr Res 2018;84. 10.1038/s41390-018-0169-7. [DOI] [PubMed] [Google Scholar]
- 2.Volpe JJ. Neonatal encephalopathy: An inadequate term for hypoxic-ischemic encephalopathy. Ann Neurol 2012;72:156–66. 10.1002/ana.23647. [DOI] [PubMed] [Google Scholar]
- 3.Kurinczuk JJ, White-Koning M, Badawi N. Epidemiology of neonatal encephalopathy and hypoxic – ischaemic encephalopathy. Early Hum Dev 2010;86:329–38. 10.1016/j.earlhumdev.2010.05.010. [DOI] [PubMed] [Google Scholar]
- 4.Nelson K, Bingham P, Edwards E, Horbar J, Kenny J, Inder T, et al. Antecedents of neonatal encephalopathy in the Vermont Oxford Network Encephalopathy Registry. Pediatrics 2012;130:878–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Jacobs S, Berg M, Hunt R, Tarnow-Mordi W, Inder T, Davis P. Cooling for newborns with hypoxic ischaemic encephalopathy. Cochrane Database Syst Rev 2013:CD003311. [DOI] [PubMed] [Google Scholar]
- 6.Shankaran S, Laptook AR, Pappas A, McDonald SA, Das A, Tyson JE, et al. Effect of depth and duration of cooling on death or disability at age 18 months among neonates with hypoxic-ischemic encephalopathy a randomized clinical trial. JAMA - J Am Med Assoc 2017;318:57–67. 10.1001/jama.2017.7218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gluckman PD, Wyatt JS, Azzopardi D, Ballard R, Edwards AD, Ferriero DM, et al. Selective head cooling with mild systemic hypothermia after neonatal encephalopathy: multicentre randomised trial. Lancet 2005;365:663–70. [DOI] [PubMed] [Google Scholar]
- 8.Wyatt JS, Gluckman PD, Liu PY, Azzopardi D, Ballard R, Edwards AD, et al. Determinants of Outcomes After Head Cooling for Neonatal Encephalopathy. Pediatrics 2007;119:912–21. 10.1542/peds.2006-2839. [DOI] [PubMed] [Google Scholar]
- 9.Graham EM, Everett AD, Delpech JC, Northington FJ. Blood biomarkers for evaluation of perinatal encephalopathy: State of the art. Curr Opin Pediatr 2018;30:199–203. 10.1097/MOP.0000000000000591. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.American College of Obstetricians and Gynecologists’ Task Force on Neonatal Encephalopathy. Neonatal encephalopathy and neurologic outcome, second edition. Obstet Gynecol 2014;123:896–901. 10.1542/peds.2014-0724. [DOI] [PubMed] [Google Scholar]
- 11.Blennow K A Review of Fluid Biomarkers for Alzheimer’s Disease: Moving from CSF to Blood. Neurol Ther 2017;6:15–24. 10.1007/s40120-017-0073-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.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. 10.1089/neu.2004.21.1553. [DOI] [PubMed] [Google Scholar]
- 13.Missler U, Wiesmann M, Wittmann G, Magerkurth O, Hagenström H. Measurement of Glial Fibrillary Acidic Protein in Human Blood: Analytical Method and Preliminary Clinical Results. Clin Chem 1999;44:138–41. [PubMed] [Google Scholar]
- 14.Ramaswamy V, Horton J, Vandermeer B, Buscemi N, Miller S, Yager J. Systematic Review of Biomarkers of Brain Injury in Term Neonatal Encephalopathy. Pediatr Neurol 2009;40:215–26. 10.1016/j.pediatrneurol.2008.09.026. [DOI] [PubMed] [Google Scholar]
- 15.Ennen CS, Huisman TAGM, Savage WJ, Northington FJ, Jennings JM, Everett AD, et al. Glial fibrillary acidic protein as a biomarker for neonatal hypoxic-ischemic encephalopathy treated with whole-body cooling. Am J Obstet Gynecol 2011;205:251.e1–251.e7. 10.1016/j.ajog.2011.06.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Massaro AN, Jeromin A, Kadom N, Vezina G, Hayes RL, Wang KKW, et al. Serum biomarkers of MRI brain injury in neonatal hypoxic ischemic encephalopathy treated with whole-body hypothermia: A pilot study. Pediatr Crit Care Med 2013;14:310–7. 10.1097/PCC.0b013e3182720642. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Chalak LF, Sánchez PJ, Adams-Huet B, Laptook AR, Heyne RJ, Rosenfeld CR. Biomarkers for severity of neonatal hypoxic-ischemic encephalopathy and outcomes in newborns receiving hypothermia therapy. J Pediatr 2014;164. 10.1016/j.jpeds.2013.10.067. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Stewart A, Tekes A, Huisman TAGM, Jennings JM, Allen MC, Northington FJ, et al. Glial fibrillary acidic protein as a biomarker for periventricular white matter injury. Am J Obstet Gynecol 2013;209:27.e1–27.e7. 10.1016/j.ajog.2013.02.049. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Represa A, Deloulme JC, Sensenbrenner M, Ben-Ari Y, Baudier J, Y B-A, et al. Neurogranin: immunocytochemical localization of a brain-specific protein kinase C substrate. J Neurosci 1990;10:3782–92. 10.1523/JNEUROSCI.10-12-03782.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Guadaño-Ferraz A, Viñuela A, Oeding G, Bernal J, Rausell E. RC3/neurogranin is expressed in pyramidal neurons of motor and somatosensory cortex in normal and denervated monkeys. J Comp Neurol 2005;493:554–70. 10.1002/cne.20774. [DOI] [PubMed] [Google Scholar]
- 21.Yang J, Korley FK, Dai M, Everett AD. Serum neurogranin measurement as a biomarker of acute traumatic brain injury. Clin Biochem 2015;48:843–8. 10.1016/j.clinbiochem.2015.05.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Çevik S, Özgenç MM, Güneyk A, Evran Ş, Akkaya E, Çalış F, et al. NRGN, S100B and GFAP levels are significantly increased in patients with structural lesions resulting from mild traumatic brain injuries. Clin Neurol Neurosurg 2019;183:105380. 10.1016/j.clineuro.2019.105380. [DOI] [PubMed] [Google Scholar]
- 23.Toorell H, Zetterberg H, Blennow K, Sävman K, Hagberg H. Increase of neuronal injury markers Tau and neurofilament light proteins in umbilical blood after intrapartum asphyxia. J Matern Neonatal Med 2017;31:1–5. 10.1080/14767058.2017.1344964. [DOI] [PubMed] [Google Scholar]
- 24.Massaro AN, Wu YW, Bammler TK, Comstock B, Mathur A, McKinstry RC, et al. Plasma Biomarkers of Brain Injury in Neonatal Hypoxic-Ischemic Encephalopathy. J Pediatr 2018;194:67–75. [DOI] [PubMed] [Google Scholar]
- 25.Takahashi K, Hasegawa S, Maeba S, Fukunaga S, Motoyama M, Hamano H, et al. Serum tau protein level serves as a predictive factor for neurological prognosis in neonatal asphyxia. Brain Dev 2014;36:670–5. 10.1016/j.braindev.2013.10.007. [DOI] [PubMed] [Google Scholar]
- 26.Lv H, Wang Q, Wu S, Yang L, Ren P, Yang Y, et al. Neonatal hypoxic ischemic encephalopathy-related biomarkers in serum and cerebrospinal fluid. Clin Chim Acta 2015;450:282–97. 10.1016/j.cca.2015.08.021. [DOI] [PubMed] [Google Scholar]
- 27.Çelik Y, Atlcl A, Gülaşl S, Makharoblldze K, Eskandari G, Sungur MA, et al. The effects of selective head cooling versus whole-body cooling on some neural and inflammatory biomarkers: A randomized controlled pilot study. Ital J Pediatr 2015;41:1–8. 10.1186/s13052-015-0188-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Chaparro-Huerta V, Flores-Soto ME, Merin Sigala ME, Barrera de León JC, Lemus-Varela M de L, Torres-Mendoza BM de G, et al. Proinflammatory Cytokines, Enolase and S-100 as Early Biochemical Indicators of Hypoxic-Ischemic Encephalopathy Following Perinatal Asphyxia in Newborns. Pediatr Neonatol 2017;58:70–6. 10.1016/j.pedneo.2016.05.001. [DOI] [PubMed] [Google Scholar]
- 29.Leifsdottir K, Mehmet H, Eksborg S, Herlenius E. Fas-ligand and interleukin-6 in the cerebrospinal fluid are early predictors of hypoxic-ischemic encephalopathy and long-term outcomes after birth asphyxia in term infants. J Neuroinflammation 2018;15:1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Imam SS, Gad GI, Atef SH, Shawky MA. Cord blood brain derived neurotrophic factor: Diagnostic and prognostic marker in fullterm newborns with perinatal asphyxia. Pakistan J Biol Sci 2009;12:1498–504. [DOI] [PubMed] [Google Scholar]
- 31.Liu F, Yang S, Du Z, Guo Z. Dynamic Changes of Cerebral-Specific Proteins in Full-Term Newborns with Hypoxic – Ischemic Encephalopathy. Cell Biochem Biophys 2013;66:389–96. 10.1007/s12013-012-9478-3. [DOI] [PubMed] [Google Scholar]
- 32.Sweetman DU, Onwuneme C, Watson WR, Murphy JFA, Molloy EJ. Perinatal Asphyxia and Erythropoietin and VEGF: Serial Serum and Cerebrospinal Fluid Responses. Neonatology 2017;111:253–9. 10.1159/000448702. [DOI] [PubMed] [Google Scholar]
- 33.O’Hare FM, Watson RWG, O’Neill A, Segurado R, Sweetman D, Downey P, et al. Serial cytokine alterations and abnormal neuroimaging in newborn infants with encephalopathy. Acta Paediatr Int J Paediatr 2017;106:561–7. 10.1111/apa.13745. [DOI] [PubMed] [Google Scholar]
- 34.Diaz-Arrastia R, Wang KKW, Papa L, Sorani MD, Yue JK, Puccio AM, et al. Acute Biomarkers of Traumatic Brain Injury: Relationship between Plasma Levels of Ubiquitin C-Terminal Hydrolase-L1 and Glial Fibrillary Acidic Protein. J Neurotrauma 2014;31:19–25. 10.1089/neu.2013.3040. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Graham EM, Burd I, Everett AD, Northington FJ. Blood Biomarkers for Evaluation of Perinatal Encephalopathy. Front Pharmacol 2016;7:196. 10.3389/fphar.2016.00196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Sarnat HB, Sarnat MS. Encephalopathy Fetal Distress A Clinical and. Arch Neurol 1976;33:696–705. [DOI] [PubMed] [Google Scholar]
- 37.Sweetman D, Kelly LA, Zareen Z, Nolan B, Murphy J, Boylan G, et al. Coagulation Profiles Are Associated With Early Clinical Outcomes in Neonatal Encephalopathy. Front Pediatr 2019;7:1–7. 10.3389/fped.2019.00399. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Douglas-Escobar M, Weiss MD. Hypoxic-Ischemic Encephalopathy A Review for the Clinician. JAMA Pediatr 2015;169:397–403. 10.1001/jamapediatrics.2014.3269. [DOI] [PubMed] [Google Scholar]
- 39.Barkovich AJ, Hajnal BL, Vigneron D, Sola A, Partridge JC, Allen F, et al. Prediction of neuromotor outcome in perinatal asphyxia: Evaluation of MR scoring systems. Am J Neuroradiol 1998;19:143–9. [PMC free article] [PubMed] [Google Scholar]
- 40.Bayley N Bayley Scales of Infant and Toddler Development Manual. 3rd edn. San Antonio, TX: The Pscyhological Corporation. 2006. [Google Scholar]
- 41.Shankaran S, Laptook AR, Ehrenkranz RA, Tyson JE, McDonald SA, Donovan EF, et al. Whole-body hypothermia for neonates with hypoxic-ischemic encephalopathy. N Engl J Med 2005;353:1574–84. 10.1056/NEJMcps050929. [DOI] [PubMed] [Google Scholar]
- 42.Tobin J Estimation of Relationships for Limited Dependent Variables. Econometrica 1958;26:24–36. [Google Scholar]
- 43.Bembea MM, Savage W, Strouse JJ, Schwartz JME, Graham E, Thompson CB, et al. Glial fibrillary acidic protein as a brain injury biomarker in children undergoing extracorporeal membrane oxygenation. Pediatr Crit Care Med 2011;12:572–9. 10.1097/PCC.0b013e3181fe3ec7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Okazaki K, Kusaka T, Kondo M, Kozawa K, Yoshizumi M, Kimura H. Temporal alteration of serum G-CSF and VEGF levels in perinatal asphyxia treated with head cooling. Cytokine 2012;60:812–4. 10.1016/j.cyto.2012.08.001. [DOI] [PubMed] [Google Scholar]
- 45.Korley FK, Diaz-Arrastia R, Wu AHB, Yue JK, Manley GT, Sair HI, et al. Circulating Brain-Derived Neurotrophic Factor Has Diagnostic and Prognostic Value in Traumatic Brain Injury. J Neurotrauma 2016;33:215–25. 10.1089/neu.2015.3949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Failla MD, Conley YP, Wagner AK. Brain-Derived Neurotrophic Factor (BDNF) in Traumatic Brain Injury-Related Mortality: Interrelationships between Genetics and Acute Systemic and Central Nervous System BDNF Profiles. Neurorehabil Neural Repair 2016;30:83–93. 10.1177/1545968315586465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Wyrobek J, Laflam A, Max L, Tian J, Neufeld KJ, Kebaish KM, et al. Association of intraoperative changes in brain-derived neurotrophic factor and postoperative delirium in older adults. Br J Anaesth 2017;119:324–32. 10.1093/bja/aex103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Olin A, Henckel E, Chen Y, Lakshmikanth T, Pou C, Mikes J, et al. Stereotypic Immune System Development in Newborn Children. Cell 2018;174:1277–1292.e14. 10.1016/j.cell.2018.06.045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Wu YW, Goodman AM, Chang T, Mulkey SB, Gonzalez FF, Mayock DE, et al. Placental pathology and neonatal brain MRI in a randomized trial of erythropoietin for hypoxic – ischemic encephalopathy. Pediatr Res 2019:1–6. 10.1038/s41390-019-0493-6. [DOI] [PubMed] [Google Scholar]
- 50.Dingley AJ, Tooley J. Xenon Ventilation During Therapeutic Hypothermia in Neonatal Encephalopathy: A Feasibility Study. Pediatrics 2014;133:809–18. [DOI] [PubMed] [Google Scholar]
- 51.Cotten CM, Murtha AP, Goldberg RN, Grotegut CA, Smith PB, Goldstein RF, et al. Feasibility of Autologous Cord Blood Cells for Infants with Hypoxic-Ischemic Encephalopathy. J Pediatr 2014;164:973–979.e1. 10.1016/j.jpeds.2013.11.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Azzopardi D, Robertson NJ, Bainbridge A, Cady E, Charles-edwards G, Deierl A, et al. Moderate hypothermia within 6 h of birth plus inhaled xenon versus moderate hypothermia alone after birth asphyxia (TOBY-Xe): a proof-of-concept , open-label , randomised controlled trial. Lancet Neurol 2016;15:145–53. 10.1016/S1474-4422(15)00347-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Leal G, Comprido D, Duarte CB. BDNF-induced local protein synthesis and synaptic plasticity. Neuropharmacology 2014;76:639–56. 10.1016/j.neuropharm.2013.04.005. [DOI] [PubMed] [Google Scholar]
- 54.Korte M, Carrolltt P, Wolf E, Brem G, Thoenent H, Bonhoeffer T. Hippocampal long-term potentiation is impaired in mice lacking brain-derived neurotrophic factor. Proc Natl Acad Sci U S A 1995;92:8856–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Grad S, Ertel W, Keel M, Infanger M, Vonderschmitt DJ, Maly FE. Strongly enhanced serum levels of vascular endothelial growth factor (VEGF) after polytrauma and burn. Clin Chem Lab Med 1998;36:379–83. [DOI] [PubMed] [Google Scholar]
- 56.Blennow M, Savman K, Ilves P, Thoresen M, Rosengren L. Brain-specific proteins in the cerebrospinal fluid of severely asphyxiated newborn infants. Acta Paediatr 2001;90:1171–5. [DOI] [PubMed] [Google Scholar]
- 57.Blennow M, Hagberg H, Rosengren L. Glial fibrillary acidic protein in the cerebrospinal fluid: A possible indicator of prognosis in full-term asphyxiated newborn infants? Pediatr Res 1995;37:260–4. 10.1203/00006450-199503000-00002. [DOI] [PubMed] [Google Scholar]
- 58.Pinchefsky EF, Hahn CD. Outcomes following electrographic seizures and electrographic status epilepticus in the pediatric and neonatal ICUs. Curr Opin Neurol 2017;30:156–64. 10.1097/WCO.0000000000000425. [DOI] [PubMed] [Google Scholar]
- 59.Numis AL, Foster-Barber A, Deng X, Rogers EE, Barkovich AJ, Ferriero DM, et al. Early changes in pro-inflammatory cytokine levels in neonates with encephalopathy are associated with remote epilepsy. Pediatr Res 2019:1–6. 10.1038/s41390-019-0473-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Binder LI, Frankfurter A, Rebhun LI. The Distribution of Tau in the Mammalian Central Nervous System. J Cell Biol 1985;101:1371–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Gorath M, Stahnke T, Mronga T, Goldbaum O, Richter-Landsberg C. Developmental Changes of Tau Protein and mRNA in Cultured Rat. Glia 2001;36:89–101. 10.1002/glia.1098. [DOI] [PubMed] [Google Scholar]
- 62.Hinojosa-Rodríguez M, Harmony T, Carrillo-Prado C, Van Horn JD, Irimia A, Torgerson C, et al. Clinical neuroimaging in the preterm infant: Diagnosis and prognosis. NeuroImage Clin 2017;16:355–68. 10.1016/j.nicl.2017.08.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63.Barkovich AJ, Miller SP, Bartha A, Newton N, Hamrick SEG, Mukherjee P, et al. MR imaging, MR spectroscopy, and diffusion tensor imaging of sequential studies in neonates with encephalopathy. Am J Neuroradiol 2006;27:533–47. [PMC free article] [PubMed] [Google Scholar]
- 64.McKinstry RC, Miller JH, Snyder AZ, Mathur A, Schefft GL, Almli CR, et al. A prospective, longitudinal diffusion tensor imaging study of brain injury in newborns. Neurology 2002;59:824–33. 10.1212/WNL.59.6.824. [DOI] [PubMed] [Google Scholar]
- 65.Bartha AI, Foster-Barber A, Miller SP, Vigneron DB, Glidden DV., Barkovich AJ, et al. Neonatal encephalopathy: Association of cytokines with MR spectroscopy and outcome. Pediatr Res 2004;56:960–6. 10.1203/01.PDR.0000144819.45689.BB. [DOI] [PubMed] [Google Scholar]
- 66.Savman K, Blennow M, Gustafson K, Tarkowski E, Hagberg H. Cytokine response in cerebrospinal fluid after birth asphyxia. Pediatr Res 1998;43:746–51. [DOI] [PubMed] [Google Scholar]