Abstract
Background and Purpose:
Preterm neonates with intraventricular hemorrhage (IVH) are at risk for post-hemorrhagic hydrocephalus and poor neurologic outcomes. Iron has been implicated in ventriculomegaly, hippocampal injury, and poor outcomes following IVH. We hypothesized that levels of cerebrospinal fluid blood breakdown products and endogenous iron clearance proteins in neonates with IVH differ from those of neonates with IVH who subsequently develop post-hemorrhagic hydrocephalus.
Methods:
Premature neonates with an estimated gestational age at birth < 30 weeks who underwent lumbar puncture for clinical evaluation an average of 2 weeks after birth were evaluated. Groups consisted of controls (n=16), low grade IVH (Grades I-II; n=4), high grade IVH (Grades III-IV; n=6) and post-hemorrhagic hydrocephalus (n=9). Control subjects were pre-term neonates born at < 30 weeks’ gestation without brain abnormality or hemorrhage on cranial ultrasound, who underwent lumbar puncture for clinical purposes. Cerebrospinal fluid hemoglobin, total bilirubin, total iron, ferritin, ceruloplasmin, transferrin, haptoglobin and hemopexin were quantified.
Results:
Cerebrospinal fluid hemoglobin levels were increased in post-hemorrhagic hydrocephalus compared to high grade IVH (9.45 vs. 6.06 μg/ml, p<0.05) and cerebrospinal fluid ferritin levels were increased in post-hemorrhagic hydrocephalus compared to controls (511.33 vs. 67.08, p<0.01). No significant group differences existed for the other cerebrospinal fluid blood-breakdown and iron-handling proteins tested. We observed positive correlations between ventricular enlargement (frontal occipital horn ratio) and ferritin (Pearson r=0.67), hemoglobin (Pearson r=0.68), and total bilirubin (Pearson r=0.69).
Conclusions:
Neonates with post-hemorrhagic hydrocephalus had significantly higher levels of hemoglobin than those with high grade IVH. Levels of blood breakdown products, hemoglobin, ferritin and bilirubin correlated with ventricular size. There was no elevation of several iron-scavenging proteins in cerebrospinal fluid in neonates with post-hemorrhagic hydrocpehalus, indicative of post-hemorrhagic hydrocephalus as a disease state occurring when endogenous iron clearance mechanisms are overwhelmed.
Keywords: intraventricular hemorrhage, germinal matrix hemorrhage, post-hemorrhagic hydrocephalus, cerebrospinal fluid, lumbar puncture, iron, biomarker
Introduction
Germinal matrix hemorrhage-intraventricular hemorrhage (GMH-IVH) is a significant cause of morbidity and mortality in preterm infants, with 40% of neonates with high grade GMH-IVH having poor outcomes or death1,2. There are 14,000 new cases of IVH per year in the US alone and 30% of infants with high grade IVH will go on to develop post-hemorrhagic hydrocephalus1, requiring lifelong treatment and associated with some of the worst neurodevelopmental outcomes in preterm infants2. There is currently no standard treatment for IVH or for the prevention of post-hemorrhagic hydrocephalus, in part as the mechanisms by which IVH results in post-hemorrhagic hydrocephalus is unknown.
Red blood cell lysis after IVH results in the release of blood breakdown products (Hb, iron, bilirubin) which have been implicated in the pathophysiology of post-hemorrhagic hydrocephalus3. Specifically, intraventricular injection of hemoglobin results in hydrocephalus in neonatal rats and this ventricular enlargement is improved with treatment of the iron chelator deferoxamine4. Additional animal studies in both neonatal and adult rodents further support hemoglobin and iron in the pathogenesis of hydrocephalus after IVH4–8. There are limited human studies evaluating blood breakdown products in the cerebrospinal fluid (CSF) of neonates and how they relate to the development of hydrocephalus9,10. To the best of our knowledge, no clinical studies which have specifically evaluated blood breakdown products and iron handling/scavenging proteins in the pathogenesis of post-hemorrhagic hydrocephalus in preterm infants. The aim of the current study is to identify endogenous iron clearance proteins and blood breakdown products in neonatal lumbar puncture (LP) CSF as associated with post-hemorrhagic hydrocephalus, and discriminating between post-hemorrhagic hydrocephalus and IVH neonates.
Materials and Methods
Study Participants
Approval to conduct the study was obtained from the Human Research Protection Office at Washington University in Saint Louis prior to the commencement of the study. Informed consent was obtained in accordance to the approved Washington University in Saint Louis Human Research Protection Office guidelines, prior to collection of clinical data, imaging and CSF information. All participant information was de-identified, coded and the identifiers or keys stored in separate files. Because of the sensitive nature of the data collected for this study, requests to access the dataset from qualified researchers trained in human subject confidentiality protocols may be sent to Washington University St. Louis at strahlej@wustl.edu. According to standard guidelines for neonatal care, all participants born ≤ 35 weeks post-menstrual age were admitted to the St. Louis Children’s Hospital Neonatal Intensive Care Unit. In the 10 year interval from 2009–2018 there were 1453 admissions to the NICU at St. Louis Children’s Hospital of neonates < 30 weeks gestation. Of those, 261 had a grade III or grade IV IVH.
Lumbar puncture CSF was collected between August 2008 and February 2017. Lumbar puncture samples were collected at the Neonatal Intensive Care Unit as part of clinical care either as part of diagnostic procedures to screen for sepsis/meningitis (n=25) or as a temporizing measure to alleviate intracranial pressure (n=3) or both (n=3). All CSF lumbar puncture samples were monitored for culture growth for a period of 3.7 ± 0.1 days, were verified to be sterile as evidenced by microbiological screening reports.
The cross-sectional study cohort included control neonates (n=16), neonates with low grade IVH (Grades I-II; n=4), neonates with high grade IVH (Grades III-IV; n=6) and neonates with post-hemorrhagic hydrocephalus (n=9). Inclusion criteria for the study were i) live-born pre-term infants with a post-menstrual age at birth < 30 weeks, with no documented congenital brain anomaly and normal ventricles without hemorrhage on cranial ultrasound who underwent lumbar puncture as part of diagnostic procedures to screen for sepsis/meningitis (control subjects); ii) neonates with low grade IVH or iii) high grade IVH, graded based on Papile’s diagnostic criteria11; iv) neonates with post-hemorrhagic hydrocephalus – IVH diagnosed on cranial imaging and a subsequent clinical diagnosis of hydrocephalus based on at least two clinical features in addition to and radiographic features of hydrocephalus: progressive increase in occipital-frontal circumference, bulging anterior fontanelle12, and/or split sutures ≥ 2 mm in the mid-parietal region (clinical features); and progressive ventriculomegaly with frontal-occipital horn ratio (FOHR) ≥ 0.5513 (radiographic feature), necessitating neurosurgical treatment for CSF diversion by 24 months age. Neurosurgical treatment for all infants with post-hemorrhagic hydrocephalus in this study was with placement of ventriculoperitoneal shunt. IVH grade was determined by radiology report (all cranial ultrasounds read by a board-certified Pediatric Radiologist) and verified by Pediatric Neurosurgeon attending blinded to the CSF results. Exclusion criteria for all participants were i) PMA birth > 30 weeks; ii) neurological diagnosis other than IVH or post-hemorrhagic hydrocephalus; iii) infants with systemic infections or positive cultures from lumbar puncture CSF, iii) evidence of congenital hydrocephalus on diagnostic imaging consistent with exam findings at birth. Limited clinical and socio-economic data were available for control subjects.
Specimen Processing
Lumbar puncture CSF samples were transported directly to the St. Louis Children’s Hospital clinical laboratory where they were frozen at −80C for a maximum of 2 days. They were then transported on ice to the Washington University Neonatal CSF Repository, where they were stored at −80C. Before experimental analysis, samples were thawed and centrifuged at 2500 rpm for 6 minutes, and the supernatant was analyzed for protein concentrations.
Quantification of CSF proteins
Quantitative measurements of Hb, total bilirubin, total iron, ferritin, ceruloplasmin, transferrin, haptoglobin and hemopexin were conducted using commercially available ELISA assays (Supplementary Table I). Assays were conducted in accordance with the manufacturer’s protocol. CSF samples and protein standards were distributed in duplicate on the assay plate. In order to overcome matrix effects and improve signal to noise ratio, we piloted assays using a series of control and post-hemorrhagic hydrocpehalus CSF samples diluted in the supplied assay diluents. CSF dilutions (Supplementary Table I) used for the assays were optimized based on the optical density measurements that fell within the range of the standard curve for both control and post-hemorrhagic hydrocpehalus pilot assays. Total protein concentration of CSF samples, diluted 1:4, were measured using the Pierce Bicinchoninic acid (Thermo Fisher Scientific, Waltham, MA). As suggested by the manufacturers, the concentrations of individual analytes were done using a 4-parameter logistic standard curve except for total iron and total protein, those were done using a linear standard curve. Where additional participant data was added to existing total protein data, inter-plate assay variability checks were not performed.
cUS reporting
Neonatal Intensive Care Units cranial ultrasounds using a Zonarez.one UltraSmartCart diagnostic system (Mountain View, CA) were acquired 1–2 times weekly for all pre-term infants to monitor for development of IVH/post-hemorrhagic hydrocpehalus and other neurological injuries.
FOHR measurements
Cranial ultrasounds (0 ± 4.10 days) closest to post-menstrual age of the sample were digitized and measurements of bi-frontal horn width (A), bi-occipital horn width (B) along with interparietal diameter (C) were determined. Estimations of ventricular size were calculated using the FOHR [FOHR = (A+B)/2C]13.
Statistical analysis
Data were analyzed by GraphPad Prism software, version 7.0 (La Jolla, CA). Group comparisons for categorical variables such as sex and race were tested with Fisher’s exact test. One-way analysis of variance (ANOVA) with Tukey’s Post-hoc test was used for testing group differences. Correlation analysis between two groups of continuous variables was determined using Pearson correlation. Receiver Operating Characteristic (ROC) curves were performed to determine ideal cut-offs for CSF hemoglobin and ferritin to discriminate between PHH post-hemorrhagic hydrocephalus and no post-hemorrhagic hydrocephalus PHH (control, grade I-II IVH, and grade III-IV IVH). Diagnostic cutoff values for prediction association with high risk of post-hemorrhagic hydrocephalus were selected based on highest Youden index (sensitivity + specificity −1). For each comparison or analysis, P value(s), followed by the result of the specific statistical test is listed. All tests were 2-tailed, reported at 95% CI and p<0.05 was considered statistically significant.
Results
Demographics and Clinical information of the study participants
CSF was analyzed from 35 participants. Demographic characteristics and mean values as well as mean CSF analyte measurements are shown in Table 1. No significant group differences in sex, race, estimated gestational age at birth, birth weight, post-menstrual age at time of sample lumbar puncture CSF collection, or in time from IVH diagnosis to lumbar puncture CSF collection were observed. As the volume of CSF required for total iron assay was much greater than that required for the remaining assays, fewer IVH and post-hemorrhagic hydrocephalus participants had CSF total iron assay completed (relative to other assays) and 5 control participants had only total iron and total protein assay completed. Numbers of participants for each CSF analyte completed are shown in Supplementary Table II.
Table 1.
Demographics and baseline clinical characteristics of participants
| Control | IVH-L | IVH-H | PHH | All | P | |
|---|---|---|---|---|---|---|
| Demographics | ||||||
| Number, n | 16 | 4 | 6 | 9 | 35 | NA |
| Male/Female, n | 8/12 | 2/2 | 6/0 | 6/3 | 26/9 | NS* |
| Caucasian/Black, n | 8/8 | 2/2 | 3/3 | 3/6 | 14/15 | NS* |
| Clinical Characteristics: Mean (SD) | ||||||
| EGA Birth, w44 | 27.1 (1.6) | 27.6 (1.6) | 26.3 (1.4) | 26.4 (2.0) | 26.8 (1.7) | NS |
| EGA Birth Weight, g | 1177.2 (288.3) | 872.5 (243.9) | 862.5 (144.8) | 908.9 (233.9) | 973.7 (259.6) | NS |
| EGA Sample, w | 28.4 (1.5) | 29.1 (3.1) | 28.4 (2.2) | 28.0 (1.7) | 28.4 (1.8) | NS |
| Time - IVH to LP, w | N/A | 2.1 (1.9) | 0.8 (0.6) | 1.7 (1.4) | 1.5 (1.4) | NS |
| Lumbar Puncture CSF Analyte Measurements: Mean (SD) | ||||||
| Hemoglobin (μg/ml) | 1.52 (1.84) | 1.31 (1.50) | 6.06 (3.32) | 9.45 (4.13) | 4.62 (4.49) | <0.0001 |
| Total Bilirubin (mg/dl) | 1.54 (0.05) | 1.71 (0.22) | 8.47 (9.31) | 9.52 (7.95) | 5.27 (6.65) | NS |
| Total Iron (μg/ml) | 0.51 (0.20) | 0.43 (0.02) | 5.29 (6.68) | 10.35 (9.70) | 3.75 (6.34) | NS |
| Ferritin (ng/ml) | 67.08 (101.80) | 315.63 (302.70) | 434.67 (215.12) | 511.33 (256.65) | 309.80 (275.19) | 0.0021 |
| Ceruloplasmin (μg/ml) | 6.57 (3.53) | 6.12 (2.00) | 5.88 (3.50) | 15.05 (16.87) | 8.78 (9.85) | NS |
| Transferrin (μg/ml) | 144.27 (53.29) | 82.30 (20.86) | 111.71 (28.64) | 547.33 (916.47) | 239.40 (499.90) | NS |
| Haptoglobin (μg/ml) | 15.57 (21.76) | 1.45 (0.99) | 1.04 (0.32) | 2.88 (2.68) | 6.49 (13.82) | NS |
| Hemopexin (μg/ml) | 34.42 (15.30) | 25.36 (9.75) | 15.28 (4.70) | 34.33 (22.58) | 29.18 (16.94) | NS |
| Total Protein (mg/dl) | 183.83 (105.5) | 148.71 (37.59) | 218.27 (46.97) | 325.42 (243.28) | 219.09 (148.40) | NS |
Abbreviations: IVH-L, Low Grade IVH (Grades I-II); IVH-H, High Grade IVH (Grades III-IV); PHH, Post Hemorrhagic Hydrocephalus; NA, Not Applicable; NS, Not Significant; EGA, Estimated Gestational Age, CSF, Cerebrospinal Fluid.
Comparison with Fisher’s Exact test (categorical variables), all comparisons for continuous variables with one-way analysis of variance (ANOVA).
CSF protein measurements
Blood and blood breakdown products:
CSF Hb levels were significantly higher in post-hemorrhagic hydrocephalus (n=8) participants compared to controls (n=11), low grade IVH (n=4) and high grade IVH (n=6,) participants (Figure 1A). Differences in the CSF levels of total bilirubin (Figure 1B) or total iron (Figure 1C) between the groups were not significant.
Figure 1. (A-D) Comparison of CSF levels of blood break-down products by group.
CSF levels of A. hemoglobin, B. total bilirubin, C. ferritin, and D. total iron in controls, low grade IVH, high grade IVH and post-hemorrhagic hydrocephalus (PHH) participants. (One-way ANOVA with Tukey’s Post-hoc multiple comparisons test)
Iron-interacting proteins:
CSF levels of ferritin were significantly higher in post-hemorrhagic hydrocephalus (n=7) and high grade IVH (n=6) participants when compared to controls (n=9) (Figure 1D). There were no significant differences in the CSF levels of ferritin between high grade IVH and post-hemorrhagic hydrocephalus. No significant differences were observed in CSF levels of ceruloplasmin and transferrin between control, low grade IVH, high grade IVH and post-hemorrhagic hydrocephalus (Figure 2A, B).
Figure 2. (A-D) Comparison of CSF levels of iron handling proteins by group.
CSF levels of A. ceruloplasmin, B. transferrin, C. haptoglobin, and D. hemopexin in controls, low grade IVH, high grade IVH and post-hemorrhagic hydrocephalus (PHH) participants. No significant group differences observed. (One-way ANOVA)
Scavenger proteins and total protein:
There were no significant group differences in the CSF levels of haptoglobin or hemopexin between control, low grade IVH, high grade IVH and post-hemorrhagic hydrocephalus participants (Figure 2C, D). Similarly, there were no significant differences in the CSF levels of total protein between the groups tested (Table 1).
Correlations among CSF proteins
We found positive correlations between CSF hemoglobin and bilirubin, hemoglobin and ferritin, total iron and total protein, and ceruloplasmin and total protein (Supplementary Figure I). CSF ceruloplasmin significantly correlated with transferrin (n = 25, p<0.001, Pearson r = 0.92) and hemopexin (n = 28, p<0.001, Pearson r = 0.66) (Supplementary Figure II). Additionally, we observed a significant positive correlation between CSF hemopexin and transferrin and total protein and transferrin (Supplementary Figure II). Subgroup analysis revealed all correlations among CSF ceruloplasmin, transferrin and hemopexin were driven by measurements from post-hemorrhagic hydrocephalus participants (Supplementary Figure III).
Ventriculomegaly, measured as frontal-occipital horn ratio (FOHR)
As expected, FOHR, a measure of ventricular enlargement with higher values corresponding to more severe ventriculomegaly, was significantly higher in post-hemorrhagic hydrocephalus (n=8) participants compared to controls (n=11), low grade IVH (n=4), and high grade IVH (n=6) participants (Figure 3). We observed significant positive correlations between FOHR and CSF ferritin, FOHR and CSF hemoglobin, and FOHR and CSF total bilirubin, indicative of higher CSF concentrations of these blood breakdown products in patients with more severe ventricular enlargement.
Figure 3. (A-D) Ventricular size: differences between patient groups and correlation of ventricular size with CSF levels of blood-breakdown products.
A. Comparison of frontal-occipital horn ratio (FOHR, measurement of ventricular size) by group: control (n=11), low grade IVH (n=4), high grade IVH (n=6), and post-hemorrhagic hydrocephalus (PHH) (n=8), validating expected between groups differences in ventricular size. B-D. Correlation of FOHR with CSF ferritin, n=26 (B), hemoglobin, n=29 (C), and total bilirubin, n=14 (D) levels, demonstrating increased ventricular size correlated with higher levels of CSF ferritin, hemoglobin, and bilirubin.
Receiver operating characteristic (ROC) analysis
We performed ROC analysis to evaluate the ability of CSF measurements to discriminate between post-hemorrhagic hydrocephalus and non-post-hemorrhagic hydrocephalus participants (controls, low grade IVH, high grade IVH). Consistent with group differences (between control, low grade IVH, high grade IVH, and post-hemorrhagic hydrocephalus) for CSF Hb and ferritin (Figure 1), the model discriminated well between post-hemorrhagic hydrocephalus disease state and non-disease state (no post-hemorrhagic hydrocephalus, all other groups) for CSF Hb and ferritin (Figure 4). Cutoff values for association of high risk of PHH was selected at 6.5 μg/ml for CSF hemoglobin (sensitivity 0.88, specificity 0.81, Youden index 0.68) and 555ng/ml for CSF ferritin (sensitivity 0.71, specificity 0.89, Youden index 0.61).
Figure 4. (A-B) ROC curves for CSF hemoglobin and ferritin.
Accuracy of CSF hemoglobin (A) and ferritin (B) in discriminating between post-hemorrhagic hydrocephalus (PHH) and non-PHH. True positive rate (sensitivity) is plotted as a function of false positive rate (1-specificity). The area under the ROC curve for hemoglobin is 0.90 (p<0.05) and for ferritin is 0.81 (p<0.05). Diagnostic cutoff values for association with high risk of post-hemorrhagic hydrocephalus were selected at 6.5(μg/ml) for hemoglobin (sensitivity 0.88, and specificity 0.81), and 555 (ng/ml) for ferritin (sensitivity 0.71, specificity 0.89). Cutoff values were selected based on optimal Youden index.
Discussion
Key Findings
We observed elevated lumbar puncture CSF hemoglobin levels in neonates who go on to develop post-hemorrhagic hydrocephalus compared to those with high grade IVH. In addition, we found elevated CSF ferritin levels in patients with post-hemorrhagic hydrocephalus compared to all other groups (control, low grade IVH and high grade IVH). We also observed an association of ventriculomegaly, as measured by FOHR on cranial US, with CSF ferritin, hemoglobin and total bilirubin levels, with higher levels of each associated with more severe ventriculomegaly. Additionally, we confirmed that radiologic measurement of ventriculomegaly (FOHR) discriminated appropriately between post-hemorrhagic hydrocephalus, IVH groups, and controls. Taken together, these findings suggest that higher levels of CSF hemoglobin and ferritin are associated with the post-hemorrhagic hydrocephalus disease state, as well as more severe degree of ventriculomegaly. ROC curves for hemoglobin and ferritin further support the validity of CSF hemoglobin and ferritin in identifying a higher risk for subsequent post-hemorrhagic hydrocephalus in preterm neonates with severe IVH, with threshold values for discriminating post-hemorrhagic hydrocephalus from IVH without post-hemorrhagic hydrocephalus. CSF hemoglobin levels > 6.5 μg/ml, and CSF ferritin levels >555ng/ml in neonates with severe IVH are associated with higher risk of post-hemorrhagic hydrocephalus in this cohort.
Role of elevated CSF Hb and ferritin
Ours is not the first study to demonstrate elevated extracellular hemoglobin in CSF following preterm IVH. Following intraventricular hemorrhage, red blood cell lysis results in release of extracellular Hb into the CSF. Lysis of red blood cells has been shown to be a critical step leading to post-hemorrhagic hydrocephalus3. Intraventricular release of extracellular Hb has been demonstrated to result in a pro-inflammatory response in the CSF following preterm IVH14. Additionally, extracellular hemoglobin has been shown to contribute to cell death in the choroid plexus following preterm IVH9, and result in extensive periventricular white matter injury following preterm IVH15.
Ferritin, an iron storage protein, primarily localized intracellularly, has been demonstrated to be elevated in CSF in neonatal IVH16. Ferritin deposition has been observed in ependymal microglial cells at autopsy in PHH17. Pathologic periventricular and hippocampal ferritin deposition has been demonstrated in animal models of IVH and post-hemorrhagic hydrocephalus3,5,6. CSF ferritin has also been observed to be elevated in a similar disease state in adults – post-hemorrhagic hydrocephalus following aneurysmal subarachnoid hemorrhage18. We observed elevated CSF ferritin levels in neonates with post-hemorrhagic hydrocephalus, as well as a significant association between CSF ferritin levels and degree of ventriculomegaly.
While the exact pathophysiologic mechanisms by which red blood cell lysis following IVH results in the development of post-hemorrhagic hydrocephalus are not fully understood, we hypothesize that the Hb metabolism is critical. Hb is metabolized by heme oxygenases to release iron and bilirubin. Ceruloplasmin oxidizes ferrous iron which is then bound by transferrin and taken up intracellularly by ferritin. We did not observe elevated CSF bilirubin in neonates with post-hemorrhagic hydrocephalus, though we did observe a correlation between CSF bilirubin levels and severity of ventriculomegaly (FOHR). Additionally, contrasting with prior evidence of elevated iron in CSF of neonates with post-hemorrhagic ventricular dilation following preterm IVH, we did not observe significant increases in CSF iron in post-hemorrhagic hydrocephalus10. Our findings with regard to CSF bilirubin levels and CSF iron levels should be interpreted with caution as sample sizes for both assays were limited and underpowered to detect a significant difference. Unlike other CSF proteins analyzed in this study, assays for total iron and total bilirubin necessitated the use of relatively larger volume of CSF for analyte quantification, which was a major limiting factor considering lumbar puncture sample availability from very few numbers of infants across groups. Further investigation is necessary to determine if CSF iron and bilirubin levels are elevated in neonates with post-hemorrhagic hydrocephalus.
Iron-handling proteins
Beyond measuring the extent of red blood cell lysis byproducts in the CSF, we sought to evaluate the response of endogenous IVH clearance mechanisms in the neonatal brain to IVH. Following IVH and intraventricular red blood cell lysis, the CSF spaces are overwhelmed by an influx of extracellular hemoglobin and hemoglobin-degradation products. The imminent physiological response to iron release is rapid detoxification of ferrous iron to prevent iron-mediated oxidative damage. Ceruloplasmin is critical to CNS iron homeostasis by oxidizing ferrous iron to non-toxic ferric iron19,20, and has been shown to be highly expressed in the choroid plexus21. Hemopexin and haptoglobin may also play an important role in iron scavenging, though the extent to which they are upregulated in the CNS following intracerebral or intraventricular hemorrhage is uncertain9,22–24. We did not find elevations of ceruloplasmin, transferrin, haptoglobin or hemopexin in the CSF of neonates with severe IVH and post-hemorrhagic hydrocephalus. This may indicate that endogenous hemoglobin clearance mechanisms are overwhelmed and that transferrin-binding and haptoglobin and hemopexin scavenging are saturated following IVH. It is unclear if this is characteristic of the developing neonatal brain or whether there is any capability of upregulating endogenous IVH clearance mechanisms in the adult brain. However, saturation of the hemoglobin-haptoglobin binding pathway in the CSF was seen in a cohort of adult patients with subarachnoid hemorrhage25. We did identify correlations between iron-handling proteins in the CSF, notably between ceruloplasmin and transferrin, ceruloplasmin and hemopexin, hemopexin and transferrin, which appeared to have been driven by observations in the post-hemorrhagic hydrocephalus group. The clinical significance of this is unclear. While there may be cross-talk among iron handlers, the net effect was insufficient to exert observable neuroprotection.
Dose-dependent IVH injury
These findings together support the notion that hemoglobin-mediated oxidative injury, cell death and subsequent post-hemorrhagic hydrocephalus is dose-dependent relating to the degree of IVH and volume of red blood cell lysis. This is in keeping with widespread clinical observations of higher IVH grade corresponding to higher risk of post-hemorrhagic hydrocephalus1. While we did observe a degree of elevation in CSF Hb levels in neonates with IVH who did not develop post-hemorrhagic hydrocephalus, CSF Hb levels were significantly higher in those with high grade IVH who went on to develop post-hemorrhagic hydrocephalus. This may indicate that a threshold level of extracellular hemoglobin resultant from red blood cell lysis in the CSF is critical to the development of hydrocephalus.
As a downstream effect of hemoglobin metabolism, we observed elevated CSF ferritin levels in neonates with PHH, as well as a clear correlation with more severe ventriculomegaly. It is unclear if this is a result of upregulation of ferritin as an endogenous IVH clearance mechanism, or rather a reflection of ependymal injury and cell death, as ferritin is a primarily intracellular protein and ferritin deposition has been demonstrated in the ependyma following post-hemorrhagic hydrocephalus17. CSF ferritin levels have also been observed to be elevated with central nervous system inflammation and meningitis, though not to the degree we observed in post-hemorrhagic hydrocephalus26,27.
While CSF ferritin has been shown to be elevated in chronic hydrocephalus following aneurysmal subarachnoid hemorrhage18, little is known about the role of blood breakdown products and iron metabolism in chronic hydrocephalus following IVH in term neonates or adults. This is an area for future study to determine if there are similarities in the underlying pathophysiology of chronic hydrocephalus in each of these post-hemorrhagic disease states.
Study Strengths
To the best of our knowledge, this is the first study to systematically investigate CSF blood breakdown products and iron handling/scavenging proteins in post-hemorrhagic hydrocephalus in preterm infants. Additionally, the correlation of elevated lumbar puncture CSF hemoglobin and ferritin with objective radiographic measurements of ventriculomegaly further strengthens the findings observed in association with the clinical state of post-hemorrhagic hydrocephalus.
Limitations
Our study is limited by small sample size, in particular with assays for iron and bilirubin which were underpowered to detect a significant difference. Additionally, inherent to studies in human neonates, there is expected variability in timing of CSF sampling by lumbar puncture following IVH, which may contribute to variability in CSF composition as some degree of IVH clearance presumably already has occurred by the time the CSF is sampled. Further, these samples are from lumbar puncture, remote from the site of hemorrhage and it is not clear if ventricular samples would differ substantially from the present findings. Our future investigations will seek to elucidate the trends in IVH clearance blood breakdown products and iron-handling proteins from serial samples of CSF from ventricular taps in neonates with PHH. An inherent limitation to our study is the potential degradation of our antigens of interest throughout their long term storage. Although our biobanking strategies for CSF have followed global standardizations28–32, including the use of a −80°C freezer, the use of polypropylene tubes, timing and speed of spinning, and a reduction in freeze/thaw cycles, specific analysis of the antigens studied here and their durability in the CSF has not been undertaken. With that said, many neurological biomarker proteins in CSF have been shown to remain stable for up to 6 years in −80°C freezer33.
Summary/Conclusions
CSF Hb and ferritin are elevated in neonatal post-hemorrhagic hydrocephalus and associated with more severe ventriculomegaly. Furthermore, we show that CSF Hb levels are elevated in patients who go on to develop post-hemorrhagic hydrocephalus compared to those with high grade IVH who do not develop hydrocephalus. No elevation of several iron-handling proteins was observed in CSF in post-hemorrhagic hydrocephalus, indicative of post-hemorrhagic hydrocephalus as a disease state occurring when endogenous iron clearance mechanisms are overwhelmed. We propose CSF Hb and ferritin as potential diagnostic biomarkers for post-hemorrhagic hydrocephalus following preterm IVH, with CSF Hb values >6.5 μg/ml and CSF ferritin values >555 ng/ml indicative of high risk for post-hemorrhagic hydrocephalus.
Supplementary Material
Acknowledgements
We thank the study participants and their families for their contributions to aid our understanding of neonatal IVH and post-hemorrhagic hydrocephalus.
Sources of Funding
Support for this work was provided by grants from the Hydrocephalus Association 2017 Innovator Award (Strahle/Mahaney), the Neurosurgeon Research Career Development Program (K12) (Strahle), NIH NS110793 (Strahle) and the Doris Duke Foundation (Strahle). The funders had no role in the study concept and design; data collection, management, analysis, and interpretation; preparation, review, or approval of the manuscript.
Footnotes
Conflict(s) of Interest/Disclosure(s)
David Limbrick, Jr., MD, PhD reports funding for research supported by Microbot Medical Inc. for work unrelated to this submitted work. The authors have no other relevant conflict(s) of interest or disclosure(s).
References
- 1.Christian EA, Jin DL, Attenello F, Wen T, Cen S, Mack WJ, et al. Trends in hospitalization of preterm infants with intraventricular hemorrhage and hydrocephalus in the United States, 2000–2010. J Neurosurg Pediatr. 2016;17:260–269. [DOI] [PubMed] [Google Scholar]
- 2.Adams-Chapman I, Hansen NI, Stoll BJ, Higgins R, NICHD Research Network. Neurodevelopmental outcome of extremely low birth weight infants with posthemorrhagic hydrocephalus requiring shunt insertion. Pediatrics. 2008;121:e1167–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gao C, Du H, Hua Y, Keep RF, Strahle J, Xi G. Role of red blood cell lysis and iron in hydrocephalus after intraventricular hemorrhage. Journal of Cerebral Blood Flow & Metabolism. 2014;34:1070–1075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Strahle JM, Garton T, Bazzi AA, Kilaru H, Garton HJL, Maher CO, et al. Role of Hemoglobin and Iron in Hydrocephalus After Neonatal Intraventricular Hemorrhage. Neurosurgery. 2014;75:696–706. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chen Z, Gao C, Hua Y, Keep RF, Muraszko K, Xi G. Role of iron in brain injury after intraventricular hemorrhage. Stroke. 2011;42:465–470. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Chen Q, Tang J, Tan L, Guo J, Tao Y, Li L, et al. Intracerebral Hematoma Contributes to Hydrocephalus After Intraventricular Hemorrhage via Aggravating Iron Accumulation. Stroke. 2015;46:2902–2908. [DOI] [PubMed] [Google Scholar]
- 7.Meng H, Li F, Hu R, Yuan Y, Gong G, Hu S, Feng H. Deferoxamine alleviates chronic hydrocephalus after intraventricular hemorrhage through iron chelation and Wnt1/Wnt3a inhibition. Brain Research. 2015;1602:44–52. [DOI] [PubMed] [Google Scholar]
- 8.Garton T, He Y, Garton HJL, Keep RF, Xi G, Strahle JM. Hemoglobin-induced neuronal degeneration in the neonatal hippocampus after intraventricular hemorrhage. Brain Research. 2016;1635:1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gram M, Sveinsdottir S, Cinthio M, Sveinsdottir K, Hansson SR, Mörgelin M, et al. Extracellular hemoglobin - mediator of inflammation and cell death in the choroid plexus following preterm intraventricular hemorrhage. J Neuroinflammation. 2014;11:200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Savman K, Nilsson UA, Blennow M, Kjellmer I, Whitelaw A. Non-protein-bound iron is elevated in cerebrospinal fluid from preterm infants with posthemorrhagic ventricular dilatation. Pediatr Res. 2001;49:208–212. [DOI] [PubMed] [Google Scholar]
- 11.Papile LA, Burstein J, Burstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J. Pediatr 1978;92:529–534. [DOI] [PubMed] [Google Scholar]
- 12.Wellons JC, Holubkov R, Browd SR, Riva-Cambrin J, Whitehead W, Kestle J, et al. The assessment of bulging fontanel and splitting of sutures in premature infants: an interrater reliability study by the Hydrocephalus Clinical Research Network. J Neurosurg Pediatr. 2013;11:12–14. [DOI] [PubMed] [Google Scholar]
- 13.O’Hayon BB, Drake JM, Ossip MG, Tuli S, Clarke M. Frontal and occipital horn ratio: A linear estimate of ventricular size for multiple imaging modalities in pediatric hydrocephalus. Pediatr Neurosurg. 1998;29:245–249. [DOI] [PubMed] [Google Scholar]
- 14.Gram M, Sveinsdottir S, Ruscher K, Hansson SR, Cinthio M, m BK, Ley D. Hemoglobin induces inflammation after preterm intraventricular hemorrhage by methemoglobin formation. J Neuroinflammation. 2013;10:1–1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ley D, Romantsik O, Vallius S, Sveinsdottir K, Sveinsdottir S, Agyemang AA, et al. High Presence of Extracellular Hemoglobin in the Periventricular White Matter Following Preterm Intraventricular Hemorrhage. Front. Physiol 2016;7:330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hällgren R, Lundqvist G, Sedin G, Wide L. Cerebrospinal fluid ferritin in newborn infants. Elevated levels in cerebral infection or bleeding. Acta Paediatr Scand. 1982;71:97–102. [DOI] [PubMed] [Google Scholar]
- 17.Fukumizu M, Takashima S, Becker LE. Glial reaction in periventricular areas of the brainstem in fetal and neonatal posthemorrhagic hydrocephalus and congenital hydrocephalus. Brain Dev. 1996;18:40–45. [DOI] [PubMed] [Google Scholar]
- 18.Suzuki H, Muramatsu M, Tanaka K, Fujiwara H, Kojima T, Taki W. Cerebrospinal fluid ferritin in chronic hydrocephalus after aneurysmal subarachnoid hemorrhage. J. Neurol 2006;253:1170–1176. [DOI] [PubMed] [Google Scholar]
- 19.Patel BN, Dunn RJ, Jeong SY, Zhu Q, Julien J-P, David S. Ceruloplasmin regulates iron levels in the CNS and prevents free radical injury. J. Neurosci 2002;22:6578–6586. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Liu H, Hua Y, Keep RF, Xi G. Brain Ceruloplasmin Expression After Experimental Intracerebral Hemorrhage and Protection Against Iron-Induced Brain Injury. Transl Stroke Res. 2019;10:112–119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Rouault TA, Zhang D-L, Jeong SY. Brain iron homeostasis, the choroid plexus, and localization of iron transport proteins. Metab Brain Dis. 2009;24:673–684. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Garland P, Durnford AJ, Okemefuna AI, Dunbar J, Nicoll JAR, Galea J, et al. Heme-Hemopexin Scavenging Is Active in the Brain and Associates With Outcome After Subarachnoid Hemorrhage. Stroke. 2016;47:872–876. [DOI] [PubMed] [Google Scholar]
- 23.Leclerc JL, Santiago-Moreno J, Dang A, Lampert AS, Cruz PE, Rosario AM, et al. Increased brain hemopexin levels improve outcomes after intracerebral hemorrhage. J. Cereb. Blood Flow Metab 2017;38:1032–1046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Righy C, Turon R, Freitas G de, Japiassú AM, Faria Neto HC de C, Bozza M, et al. Hemoglobin metabolism by-products are associated with an inflammatory response in patients with hemorrhagic stroke. Revista Brasileira de Terapia Intensiva. 2018;30:1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Galea J, Cruickshank G, Teeling JL, Boche D, Garland P, Perry VH, Galea I. The intrathecal CD163-haptoglobin-hemoglobin scavenging system in subarachnoid hemorrhage. J. Neurochem 2012;121:785–792. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Shokrollahi MR, Shabanzadeh K, Noorbakhsh S, Tabatabaei A, Movahedi Z, Shamshiri AR. Diagnostic Value of CRP, Procalcitonin, and Ferritin Levels in Cerebrospinal Fluid of Children with Meningitis. CNSAMC. 2018;18:1–5. [DOI] [PubMed] [Google Scholar]
- 27.Sanaei Dashti A, Alizadeh S, Karimi A, Khalifeh M, Shoja SA. Diagnostic value of lactate, procalcitonin, ferritin, serum-C-reactive protein, and other biomarkers in bacterial and viral meningitis. Medicine. 2017;96:e7637–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Teunissen CE, Petzold A, Bennett JL, Berven FS, Brundin L, Comabella M, et al. A consensus protocol for the standardization of cerebrospinal fluid collection and biobanking. Neurology. 2009;73:1914–1922. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Teunissen CE, Tumani H, Engelborghs S, Mollenhauer B. Biobanking of CSF: international standardization to optimize biomarker development. Clin. Biochem 2014;47:288–292. [DOI] [PubMed] [Google Scholar]
- 30.Willemse EAJ, Teunissen CE. Biobanking of Cerebrospinal Fluid for Biomarker Analysis in Neurological Diseases. Adv. Exp. Med. Biol 2015;864:79–93. [DOI] [PubMed] [Google Scholar]
- 31.Vanderstichele H, Bibl M, Engelborghs S, Le Bastard N, Lewczuk P, Molinuevo JL, et al. Standardization of preanalytical aspects of cerebrospinal fluid biomarker testing for Alzheimer“s disease diagnosis: a consensus paper from the Alzheimer”s Biomarkers Standardization Initiative. Alzheimers Dement. 2012;8:65–73. [DOI] [PubMed] [Google Scholar]
- 32.del Campo M, Mollenhauer B, Bertolotto A, Engelborghs S, Hampel H, Simonsen AH, et al. Recommendations to standardize preanalytical confounding factors in Alzheimer“s and Parkinson”s disease cerebrospinal fluid biomarkers: an update. Biomark Med. 2012;6:419–430. [DOI] [PubMed] [Google Scholar]
- 33.Schipke CG, Jessen F, Teipel S, Luckhaus C, Wiltfang J, Esselmann H, et al. Long-term stability of Alzheimer’s disease biomarker proteins in cerebrospinal fluid. J. Alzheimers Dis 2011;26:255–262. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.