 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
ARTICLE |
a Department of Anatomy and Cell Biology, University of Melbourne, Victoria, Australia
b Department of Pediatrics, Royal Children's Hospital, Parkville, Victoria, Australia
c Department of Pediatrics, St Louis Children's Hospital, Washington University, St Louis, MO
d Division of Clinical Sciences, Imperial College, London, United Kingdom
e Departments of Pathology, University of Texas Health Science Center, San Antonio, Texas
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
METHODS. Fetal baboons were delivered at 125 days of gestation (term: 
185 days of gestation) by cesarean section. Animals were maintained for 28 days postdelivery with either: early continuous positive airway pressure (commencing at 24 hours; n = 6) or delayed continuous positive airway pressure (positive pressure ventilation for 5 days followed by nCPAP; n = 5). Gestational controls (n = 4) were delivered at 153 days of gestation. At the completion of the study, animals were killed, the brains were assessed histologically for growth and development, and evidence of cerebral injury and indices for both parameters were formulated.
RESULTS. Brain and body weights were reduced in all of the nasal continuous positive airway pressure animals compared with controls; however, the brain/body weight ratio was increased in early continuous positive airway pressure animals. Within both nasal continuous positive airway pressure groups compared with controls, there was increased gliosis in the subcortical and deep white matter and cortex and a persistence of radial glia. Early continuous positive airway pressure was associated with less cerebral injury than delayed continuous positive airway pressure therapy. Neuropathologies were not observed in controls.
CONCLUSIONS. Premature delivery, in the absence of potentiating factors, such as hypoxia or infection, is associated with a decrease in brain growth and the presence of subtle brain injury, which seems to be modified by respiratory therapies with early continuous positive airway pressure being associated with less overall cerebral injury.
Key Words: nonhuman primate • prematurity • cerebral injury • white matter injury • nasal continuous positive airway pressure • positive pressure ventilation • neuropathology • brain development
Abbreviations: nCPAP—nasal continuous positive airway pressure • EnCPAP—early nasal continuous positive airway pressure • DnCPAP—delayed nasal continuous positive airway pressure • dg—days of gestation • PPV—positive pressure ventilation • PIP—peak inspiratory pressure • PEEP—positive end-expiratory pressure • FIO2—fraction of inspired oxygen • PaO2—partial arterial pressure of oxygen • PaCO2—partial arterial pressure of carbon dioxide • H&E—hematoxylin and eosin • LFB—luxol fast blue • GFAP—glial fibrillary acidic protein • SFI—surface folding index • HR—heart rate • MAP—mean arterial pressure
Mortality rates for very low birth-weight preterm infants (<30 weeks or 1250 g) have fallen over the last few decades1 with therapies such as antenatal steroids, exogenous surfactant administration, and new ventilator strategies all having played a role.2 However, despite the improvement in mortality, long-term neurodevelopmental morbidity remains higher than desired.3,4 As many as 10% to 15% of very preterm infants are diagnosed with cerebral palsy, and 
50% exhibit motor, cognitive, and behavioral deficits, including lower intelligence quotient, attention-deficit/hyperactivity disorder, anxiety disorders, and learning disabilities.5,6
The major neuropathologies contributing to the risk of adverse neurologic outcome in the preterm infant are associated with the cerebral white matter, including periventricular leukomalacia and periventricular hemorrhagic venous infarction (grade IV intraventricular hemorrhage).7 However, it is also increasingly recognized that preterm delivery may be associated with alterations in cerebral structural development.8 Over the last decade in the preterm infant, there has been an increasing trend toward early extubation to nasal continuous positive airway pressure (nCPAP) with some evidence that the early use of nCPAP may be associated with a reduction in the need for mechanical ventilation,9–11 reduction in retinopathy of prematurity, and chronic lung disease.12,13 However, there are little data on the impact of early nCPAP on the risk of cerebral injury or alterations in cerebral growth in the preterm infant.
To investigate the impact of randomized neonatal therapies on neuropathological outcomes, an animal model of preterm birth and neonatal care is essential. Such a model has been developed in the immature baboon (Papio papio)14 and used to study the impact of early nCPAP (EnCPAP) or delayed nCPAP (DnCPAP) on respiratory outcome, with the secondary aim of assessing the consequences on the brain. We have reported previously on the relevance of the model in regard to the nature and extent of neuropathology.15 Because this model closely resembles the clinical situation of the human premature infant, it has the potential to shed important light on the effect of respiratory therapies on the nature and severity of brain injury. Thus, in this primate model (delivered prenatally at 125 days of gestation [dg]; term: 185 dg; equivalent to 26 weeks of gestation in human neonates), we aimed to assess the consequences of premature delivery managed with either an EnCPAP or DnCPAP protocol on brain growth and neuropathology.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Delivery and Instrumentation
Timed gestations were determined by observing characteristic sex skin changes and confirmed by serial fetal ultrasound examinations. Pregnant baboon dams (P papio) were treated with 6 mg of intramuscular betamethasone 48 and 24 hours before elective hysterotomy under general anesthesia. Study animals were delivered at 125 ± 2 days (67% of term gestation at 185 days). At birth animals were weighed and intubated with a 2-mm endotracheal tube. All of the animals were treated with 200 mg/kg of Curosurf (provided by Chiesi Farmaceutica SpA, Parma, Italy) before the initiation of ventilator support.
Respiratory Management
The respiratory management of both EnCPAP and DnCPAP infants was initially identical and is described in detail in the accompanying article.16 Briefly, positive pressure ventilation (PPV) was initiated with a humidified pressure-limited, time-cycled ventilator. The initial rate was 40 breaths per minute, peak inspiratory pressure (PIP) adequate to move the chest, positive end-expiratory pressure (PEEP) at 5 cm H2O, and fraction of inspired oxygen (FIO2) of 0.40. In both groups, PEEP remained constant at 5 cm H2O; PIP, FIO2, and rate were adjusted to achieve target levels of partial arterial pressure of oxygen (PaO2) at 55 to 70 mmHg, PaCO2 at 50 to 60 mmHg, and pH > 7.2. Target ventilation parameters of FIO2 <0.3, PIP 14 to 16 cm H2O, PEEP of 5 cm H2O, and a rate of 20 breaths per minute were aimed for over the first 24-hour study period in the EnCPAP group. A repeat dose of surfactant (Curosurf 100 mg/kg) was administered routinely at 6 hours of age to both groups. Animals received caffeine citrate (20 mg/kg) intravenously at 1 hour and 12 hours of age and daily thereafter (10 mg/kg), with the exception of 1 DnCPAP infant in which caffeine was given first on day 10 rather than day 1 as with other animals. Sedation was kept to a minimum, especially in the 12 hours before extubation; however, if the animal experienced distress, chloral hydrate (10–15 mg) or ketamine (2.5 mg/kg) was administered as required.
Extubation to nCPAP was attempted at either 24 hours (EnCPAP) or 5 days (120 hours) of age (DnCPAP). All of the animals were maintained on the Infant Flow Generator nCPAP delivery device (provided by EME [ElectroMedical Equipment] Ltd, Brighton, United Kingdom) via nasal prongs and occasionally nasal mask with an initial pressure of 7 cm H2O. The animal continued on nCPAP as long as there was an adequate respiratory drive; the criteria for which included a FIO2 <0.5 and a pH >7.20, with no limit set for partial pressure of carbon dioxide (PaCO2) provided the pH was maintained. If the nCPAP treatment failed, the animal was reintubated and ventilated with the least support to achieve adequate gas exchange and chest inflation as described above. If the animal had minimal oxygen requirements (FIO2 <0.25), good respiratory effort, and no chest retractions, nCPAP was discontinued and the animal placed in humidified supplementary oxygen or air. nCPAP was reinstated if inspired FIO2 exceeded 0.25 or poor respiratory effort or if chest retractions were observed. The management of nutrition, patent ductus arteriosus, and hypotension have been described previously.14
Histologic Analysis
At 151 ± 1 dg, all of the prematurely delivered animals were killed with sodium pentobarbitone (intravenous 130 mg/kg). In addition, gestational control animals (n = 4) were delivered at 153 dg by hysterotomy under general anesthesia and killed immediately with sodium pentobarbitone. Brains were removed, weighed, immersed in 10% buffered formalin, and shipped to the University of Melbourne for histologic analysis. A tissue slicer (SliceOmatic, Virtual Magic Inc, Montreal, Canada) was used to section each brain into 5-mm coronal blocks; depending on the size of the brain this ranged from 10 to 13 blocks. Blocks from the right hemisphere of each brain were processed with paraffin and 10 (8 µm) sections collected from the rostral surface of each block.
Histology
A section from each block was stained with hematoxylin and eosin (H&E) and assessed qualitatively for gross morphologic changes, including the presence of hemorrhages (subarachnoid, germinal matrix, and intraventricular), lesions or infarcts, axonal injury, and gliosis. Sections containing hemorrhages (identified with H&E staining) were stained with Perls to visualize hemosiderin deposition; depositions within hemorrhages suggest that the bleed has been present for 
48 hours. Luxol fast blue (LFB) staining was used to assess the extent of myelination. The presence of gliosis was confirmed with glial fibrillary acidic protein (GFAP) immunohistochemistry and microglia with lectin histochemistry (see below).
Immunohistochemistry
Immunohistochemistry for rabbit anti-GFAP (1:500, Sigma, St Louis, MO) was used to identify astrocytes, mouse anti-human proliferating cell nuclear antigen (1:1000, BD Biosciences, San Jose, CA) to identify mitotic cells, and rabbit anti-human activated Caspase 3 (1:1000, R&D systems Inc, Minneapolis, MN) to identify cell death using the avidin-biotin peroxidase complex (Vector Laboratories, Burlingame, CA), as described previously.15 Control and experimental material were stained simultaneously to avoid procedural variation. Control experiments were performed omitting the primary antibodies, whereupon staining failed to occur. Sections were counterstained with 0.01% thionin.
Lectin Histochemistry
Lectin histochemistry was used to visualize macrophages/reactive microglia. The same protocol was used as for GFAP immunohistochemistry, with the exception that sections were treated with 5% fish gelatin (blocking solution, Sigma Chemical Company, St Louis, MO) before incubation in biotinylated Lycopersicin esculentum (tomato) lectin for 1 night (1:250, Sigma Chemical Company), and the secondary antibody was omitted. Sections were then counterstained with 0.01% thionin.
Qualitative Analysis
H&E-stained sections were assessed for evidence of hemorrhages or overt injury, such as infarcts, cystic white matter lesions, or neuronal death. The presence of lectin-positive reactive macrophages/microglial cells was examined qualitatively in the frontal/temporal, parietal/temporal, and occipital gray and white matter.
Quantitative Analysis
All of the analyses were performed on all of the brains in the study. Measurements were made on coded slides blinded to the observer with the codes not being disclosed until the conclusion of analyses. Areas were assessed using a digitizing program (Sigma Scan Pro v4, SPSS Science, Chicago, IL) and counts preformed using an image analysis system (Image Pro Plus v4.1, Media Cybernetics, Silver City, MD)
Volumetric Measurements
In H&E-stained coronal sections, the cross-sectional area of the right hemisphere of each block was assessed. Volumes were then estimated using the Cavalieri principle.17 The total volume of white matter, deep gray matter (basal ganglia, thalamus, and hippocampus), and cortex were also assessed in this manner and were expressed as a percentage of total volume.
Surface Folding Index
The surface folding index (SFI), which gives an estimation of the expansion of the surface area relative to volume, was determined from H&E-stained sections.15 The area (A) of the entire cerebral hemisphere (excluding the deep gray matter) and the length (L) of the pial boundary of the cerebral cortex were measured (Sigma Scan Pro) at each level to calculate a mean SFI (SFI = L2/A) for each animal. SFI of the frontal/temporal, parietal/temporal, and occipital regions were calculated from these data to assess regional growth rates.
Percentage of White Matter Occupied by Blood Vessels
A point counting technique18 was used to determine the density of blood vessel profiles in randomly selected regions of deep and subcortical white matter (x660) at each brain level and a mean value calculated. Assessment was performed in GFAP-immunoreactive stained sections, because blood vessel profiles are clearly delineated.19
Areal Density of Astrocytes
For the deep and subcortical white matter, one region was selected randomly from a section of each block (x660). For the cortex, sections were selected from the frontal/temporal, parietal/temporal, and occipital lobes; 3 areas of cortex randomly sampled in each (x660; 9 in total from each animal); and a mean of means calculated. In addition, the cortex at the base of the sulcus between the superior and the middle temporal gyri was also analyzed (x660; 3 sample points from each). In the hippocampus, 2 randomly selected regions of the stratum radiatum were analyzed (x660) and a mean taken.
Semiquantitative Analysis
Myelination
In gestational control brains, myelination was most advanced in the internal capsule; this was given a score of 3. The extent of myelination in the 2 nCPAP groups was scored against this standard in sections from the frontal/temporal, parietal/temporal, and occipital regions (0: no myelination; 1: a few myelinated fibers; 2: bundles of myelinated fibers present; and 3: similar extent of myelination to gestational controls).
Perivascular Cuffing in the Subcortical White Matter
The extent of perivascular cuffing was assessed on H&E-stained sections at each level and scored on a scale of 0 to 3 (0: not observed; 1: occasionally observed; 2: moderate degree; and 3: considerable number of vessels with perivascular cuffing).
GFAP-Immunoreactive Radial Glial Fibers
Sections from the frontal/temporal, parietal/temporal, and occipital regions were examined and the presence of radial glial fibers scored on a scale of 0 to 3 (0: not observed; 1: occasionally observed; 2: moderate degree; and 3: considerable number of intensely GFAP-immunoreactive radial glial fibers observed).
GFAP-Immunoreactive Astrocytes at the Base of the Gyri
The degree of astrocytic hypertrophy was graded on a scale of 0 to 3 (0: no hypertrophy; 1: mild; 2: moderate; and 3: marked hypertrophic cell bodies and extensive fiber processes).
Growth and Development Index
To assess cerebral development in prematurely delivered baboons compared with gestational controls, a growth index was constructed including body and brain weights and the SFI. Data points were ranked from heaviest (15 points) to lightest (1 point) for weights and from highest (15 points) to lowest (1 point) for SFI. Scores were added together and a mean calculated for each group.
Brain Damage Index
To assess the extent of structural alterations observed in nCPAP compared with gestational control baboon neonates, a brain index was devised.20 This index included semiquantitative estimations of the extent of myelination, perivascular cuffing, degree of GFAP-immunoreactive astrocytic hypertrophy at the base of sulci, and presence of GFAP-immunoreactive radial glial fibers, as well as quantitative estimations of the areal density of astrocytes in the subcortical and deep white matter, cortex, and at the base of the sulcus located between the superior and middle temporal gyri. If a difference existed between control and nCPAP animals, data points were ranked from the most affected (15 points) to least affected (1 point); for identical values, the same ranking score was given to each animal. The brain index was determined by adding the scores from each category. Two observers scored parameters on coded slides. For some parameters, the frontal/temporal, parietal/temporal, and occipital regions were scored separately to determine whether a particular region of the brain was more susceptible to damage.
Physiologic Data
Physiologic data, including arterial blood gases (PaO2, PaCO2, and pH) and systolic, diastolic, and mean arterial blood pressure, were monitored via an umbilical arterial catheter until day 10. Heart rate (HR), FIO2, and type of ventilator support required were recorded throughout the experimental period. The maximum and minimum values were recorded every 6 hours for the first 2 days and daily thereafter. The overall maximum and minimum values were calculated for each parameter over the first 10 days of the ventilatory period with the exception of FIO2, which was recorded over 28 days.16 In addition, the flux, which is an indication of the variability for each parameter, was calculated by subtracting the minimum from the maximum value at each time point.
Regression Analysis
Linear regression analysis was conducted to determine whether there was a correlation between: (1) physiologic variables and the brain damage index; (2) physiologic variables and quantitative parameters (astrocyte densities and volumetric measurements); (3) quantitative parameters (astrocyte densities) and indices of brain growth; and (4) total time on PPV with brain injury. A P < .05 was considered to be significant.
Data Analysis
The statistical significance of differences between nCPAP and control groups was tested using a 1-way analysis of variance with posthoc analysis (Tukey's test) for histologic parameters and t tests to compare between nCPAP groups for physiologic parameters; a P < .05 was considered to be significant. Results are expressed as mean ± SEM (weights and areas) and mean of means ± SEM (histologic parameters).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Growth and Development
Brain and Body Weights
Body and brain weights were reduced in both nCPAP groups of animals compared with age-matched gestational controls (P < .01). The brain/body weight ratio was increased (P < .05) in EnCPAP but not DnCPAP compared with controls (Table 1).
|
SFI
The SFI of the cerebral hemispheres overall was significantly reduced (P < .01) in both groups of nCPAP animals compared with controls (60.1 ± 2.4, control vs 45.3 ± 1.6, EnCPAP and 46.9 ± 2.5, DnCPAP). On a regional basis, there was a significant reduction in the parietal/temporal region in EnCPAP (P < .001) and DnCPAP (P < .01) compared with controls (80.4 ± 4.9, control vs 51.6 ± 2.5, EnCPAP and 57.5 ± 4.7, DnCPAP) but not in other regions.
Growth and Development Index
Growth and development index parameters comprising this score are summarized in Table 2. There was a decrease in the score for both DnCPAP (P < .01) and EnCPAP (P < .001) animals compared with controls (Fig 2A).
|
|
|
Quantitative Assessment
Areal Density of Astrocytes
Deep White Matter
There was an increase in the areal density of astrocytes in the deep white matter in EnCPAP (P < .01) and DnCPAP (P < .01) animals compared with controls (Tables 3 and 4). This is illustrated by comparing Fig 1A (gestational control) with Fig 1B (DnCPAP). Within specific lobes, the densities in DnCPAP were increased in the frontal/temporal and parietal/temporal lobes (P < .01) but not in the occipital lobe, whereas in EnCPAP there was an increase in the parietal/temporal (P < .05) and occipital (P < .05) but not the frontal/temporal lobes.
|
|
Cerebral Cortex
There was an overall increase in the areal density of astrocytes in the cortex in DnCPAP (P < .01) and EnCPAP (P < .05) animals compared with controls (Table 4); this was particularly prominent in the frontal/parietal (P < .01) and occipital (P < .05) regions in DnCPAP and the frontal/parietal region (P < .01) in EnCPAP animals. The areal density of astrocytes at the base of the sulcus between the superior and middle temporal gyri was increased in DnCPAP compared with controls (P < .01) and in DnCPAP compared with EnCPAP (P < .05) animals. There was no difference (P > .05) in the areal density of astrocytes in the hippocampus between nCPAP and control animals.
Percentage of White Matter Occupied by Blood Vessels
The percentage of neuropil occupied by blood vessels was not different between groups (P > .05) in either the subcortical (control, 0.88 ± 0.11%; DnCPAP, 1.62 ± 0.27%; EnCPAP, 1.42 ± 0.26%) or deep white matter (control, 0.96 ± 0.25%; DnCPAP, 1.53 ± 0.17%; EnCPAP, 1.60 ± 0.30%) regions.
Semiquantitative Assessment
Myelination
In frontal/temporal regions of control brains, myelination (LFB-positive staining) was observed at the level of the internal capsule with sparse fine fibers extending into the deep white matter. At the parietal/temporal level, a dense fiber network was observed in the thalamus and basal ganglia with fibers projecting dorsally through the deep white matter into the tips of the postcentral and precentral gyri (Fig 1 G and H); ventrally projecting fibers were present but not as prominent at this age. Strong bands of fibers were also observed in the lateral geniculate. In the occipital region, very little myelination was observed. A semiquantitative assessment (Table 4) revealed no significant reduction in myelination across the brain in nCPAP animals compared with controls, although there was a tendency toward a reduction in the DnCPAP group.
Radial Glia
GFAP-immunoreactive radial glial fibers are rarely observed in the control brain at 153 dg. Qualitative examination revealed intensely immunoreactive radial glial fibers at the ventricular surface projecting into the deep and occasionally subcortical white matter in both groups of nCPAP animals. This is illustrated by comparing Fig 1C (control) with Fig 1D (DnCPAP). Assessment showed that there was an increased presence of GFAP-immunoreactive fibers in both DnCPAP (P < .01) and EnCPAP (P < .05) compared with control animals (Table 4). There was a tendency for this to be pronounced in the frontal region in the DnCPAP animals, but this was not significant.
Perivascular Cuffing
Perivascular cuffing was increased in the subcortical and deep white matter in nCPAP animals compared with controls (P < .05; Table 4).
Hypertrophic Astrocytes
In addition to an increase in astrocytic density in nCPAP, astrocytic hypertrophy was also observed in nCPAP animals primarily at the base of the sulcus between the superior and middle temporal gyri but in more severe cases extending to the sulci between the postcentral and superior temporal gyri. This is illustrated for an EnCPAP animal (Fig 1E control vs Fig 1F EnCPAP). There was an increase in hypertrophic astrocytes in DnCPAP compared with controls (P < .05; Table 4). There was no evidence of either mitosis (proliferation cell nuclear antigen-immunoreactive) or increased cell death (activated Caspase 3-immunoreactive) in regions of hypertrophic astrocytes.
Brain Damage Index
The parameters comprising the brain damage index score are summarized in Table 5. There was an increase in the brain damage index score for both DnCPAP (P < .001) and EnCPAP (P < .01) animals compared with controls and for DnCPAP compared with EnCPAP animals (P < .05; Table 4; Fig 2B).
|
Flux Changes
The FIO2 maximum (P < .05), median (P < .01), and average (P < .01) flux; the PaCO2 median (P < .05); and the MAP average (P < .05) flux values were increased in DnCPAP compared with EnCPAP animals. There was no difference between DnCPAP and EnCPAP groups in relation to the maximum, median or average flux values for PaO2, pH, or HR.
Relationship of Brain Injury to Brain Volumes and Physiology
There was a negative correlation between increasing white matter volume and the areal density of astrocytes in both the deep white matter (r2 = 0.41; P < .001; Fig 2C) and the subcortical white matter (r2 = 0.32; P < .05), indicating that as the number of astrocytes increases, then the volume of white matter decreases. There was a positive correlation between the presence of GFAP-immunoreactive radial glial fibers and the number of astrocytes in the deep white matter (r2 = 0.39; P < .01) and the total number of astrocytes in the white matter (r2 = 0.54; P < .002).
There was a positive correlation between the median flux in FIO2 and the brain index (r2 = 0.59; P < .006; Fig 3A), density of cortical astrocytes at the base of the sulcus (r2 = 0.42; P < .03; Fig 3B), and density of cortical astrocytes (r2 = 0.46; P < .02). There was a positive correlation between the average flux in FIO2 and GFAP-immunoreactive radial glial fibers (r2 = 0.49; P < .02). In addition, the median flux in pH also correlated with the brain damage index (r2 = 0.49; P < .02; Fig 3C). There was also a positive correlation between the maximum flux in PaCO2 and white matter astrocytes (r2 = 0.45; P < .03; Fig 3D). Thus, the greater the flux in oxygen requirement and in pH, the greater the extent of injury. There was also a trend toward a positive correlation between the total time on PPV and a higher brain damage score (P < .06).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
28 days, exhibit delayed body and brain growth, and develop cerebral neuropathologies when compared with gestational controls. Furthermore within this paradigm, maintaining PPV after delivery for 5 days rather than 1 day before weaning to nCPAP seems to increase the adverse effects on brain growth and the incidence of neuropathologies.
Brain Injury
We have shown that cerebral injury occurs in all prematurely delivered baboons in the absence of severe potentiating factors, such as perinatal infection or hypoxia. No cystic infarction in the white matter was noted in any animal, but the diffuse pattern and more subtle nature of injury is consistent with that reported in recent neuroimaging studies in the preterm infant.21–23 More importantly, in this randomly assigned series of animals, delaying the extubation to nCPAP seemed to result in a higher incidence of neuropathologies as evident by the higher brain damage index in DnCPAP compared with EnCPAP animals.
This model continued to demonstrate that the major region of vulnerability in the preterm brain lies within the cerebral white matter. Reactivity within this region was seen in both EnCPAP and DnCPAP animals with an increase (P < .01) in the areal density of astrocytes in the deep and subcortical white matter and with cuffing of capillaries in the subcortical white matter consistent with macrophage invasion from the periphery possibly in response to chemokine signals from affected neurons.24 There also seemed to be a delay in myelination in the DnCPAP animals. The persistence of radial glia in both cohorts of nCPAP animals was of particular interest; in gestational controls at 153 days, radial glia were not present in the white matter, because they seemed to have differentiated into astrocytes. A recent study in rodents has shown that chronic hypoxia in the perinatal period promotes the appearance of radial glia throughout the subventricular and ependymal zones.25 The authors hypothesized that hypoxia may lead to differentiation of astrocytes into radial glial cells, which may then divide and give rise to new neurons; radial glia are now known to be neural precursors.26 An alternative hypothesis is that hypoxia might inhibit the normal maturation of radial glia into astrocytes leading to their prolonged presence in the developing brain. Currently we are not able to distinguish between these possibilities in the premature baboon brain; we did find, however, that the greater the flux in oxygen requirements, the more pronounced was the presence of radial glia. Our data do support an association between the presence of radial glia and the presence and extent of white matter injury (as evidenced by reactive gliosis). Reactive astrocytes produce insulin-like growth factor 127 and fibroblast growth factor28 to support neuronal survival and to assist in repair mechanisms, but they also produce cytokines and reactive oxygen species which could exacerbate injury via oxidative and inflammatory pathways. The signaling factors mediating persistence or generation of the radial glia at this point in cerebral maturation are worthy of further study. The sequence of evolution of astrogliosis and the radial glia could be revealed with serial analysis at multiple time points along the course of the experiments.
In addition to glial proliferation, glial hypertrophy was present in a unique distribution at the base of the sulci, particularly between the superior and the middle temporal gyri. This was most prominent in the DnCPAP animals. The area at the base of the sulcus is known to be vulnerable to ischemic cerebral injury possibly because of the relative avascularity of this region as it is located between penetrating vessels from the meningeal arteries.29 There was no evidence of neuronal or glial apoptosis or mitosis in this region, but we cannot exclude that it could have occurred at an earlier point in the study. A similar vulnerability for the depths of the sulci in these regions has been observed with neuropathology and more recently with signal abnormalities on MRI scans in term infants with perinatal asphyxia.7 There was some regional predilection with the parietal/temporal lobes being most affected and the occipital lobe least affected. The presence of cortical glial proliferation at the base of the sulci was not a prominent finding in our initial PPV models,15 suggesting that there are exposure factors in this model predisposing to this lesion. Such a hypothesis would also have to support the greater extent of the lesion in the DnCPAP animals. This model was the only group studied to date that has received minimal sedation and routine therapeutic caffeine, and, thus, they have potentially suffered more instability in ventilatory parameters with apneic and bradycardic episodes.16 Caffeine is an adenosine receptor antagonist that has been shown to be involved in the pathway to white matter injury in a mouse model of hypoxic ventriculomegaly30 and, thus, may be neuroprotective in the white matter. However, caffeine is also a neuronal stimulant, which may upregulate metabolic demand. If this increased demand is compromised with frequent hypoxic and/or bradycardic episodes, then this watershed region will be "exposed" with increased demand and reduced supply. Such a hypothesis is supported by the observation that the DnCPAP animals were more difficult to maintain on nCPAP with more frequent reintubation for apnoea and bradycardia. Thus, caffeine may be neuroprotective in the cerebral white matter and yet potentially increase vulnerability in the cortical gray matter to intermittent hypoperfusion. Further investigation of these effects is critical, because caffeine or theophylline are in routine use in the NICU.
Brain Growth
Although the body and brain weights were reduced with both early and late extubation to nCPAP, the brain/body weight ratio was increased (P < .05) in EnCPAP animals compared with gestational controls suggesting that there was more brain sparing within this model. There was a reduction in all of the volumetric measurements in the cerebral hemispheres of the prematurely delivered animals compared with controls with the exception of the white matter in EnCPAP animals. This was reflected in an increased ratio of white matter to total hemispheric volume between this group and both gestational controls and DnCPAP animals. It seems that EnCPAP was associated with selective improvement in white matter volume, which we hypothesize results from a reduction in white matter injury with less gliosis, as well as other potentially beneficial effects of early weaning on cerebral white matter development.
Gyrification of the cerebral hemispheres is an important index of brain development and is thought to reflect development of the cerebral and subcortical connectivity. Hypotheses to account for gyrification include mechanisms intrinsic to the cortical gray matter,31 such as differential growth of different layers and/or theories around tension proposing that regions with greater connectivity provide tension that draws them together forming gyri, whereas more weakly interconnected cortical regions drift apart allowing for the formation of sulci.32 Afferent and efferent connections with subcortical structures may also influence cortical folding although to a lesser degree, because the tangential force components are weaker.32 In our study, the SFI of the hemispheres was reduced (P < .01) in both EnCPAP and DnCPAP groups indicating that cortical development was affected. We cannot specifically determine which aspects of cortical connectivity have been altered or whether there are indeed any differences between the 2 ventilatory regimes. Because advanced MRI techniques are now being used to compare brain surface morphology and measures of gyrification in infants who are born preterm,8,33 as well as in this model,34 better delineation in vivo of the development of the complexity of the cortical layers and the relationship of white matter and cortical development will be possible. It is of interest that recently MRI has been used in childhood- and adolescent-onset schizophrenia to identify gyrification abnormalities, which could reflect developmental aberrations in cerebral and subcortical connectivity.35
The question arises as to why premature delivery causes mild cerebral injury and delayed brain growth and, further, why delaying extubation to nCPAP might exacerbate this response. There are important limitations to note in the comparison of these 2 groups. The DnCPAP group had a higher ratio of males, experienced more sepsis,16 and may have experienced more physiologic instability with more apnea and bradycardia requiring reintubation.16 Some of these factors, although related to the ventilatory management, may have been more potent as independent mediators of brain injury. The major mediator of poor brain growth seems to be the presence of cerebral injury, particularly that of cerebral white matter injury with diffuse gliosis in the white matter being associated with reduced brain volume. One of the etiologies for white matter injury, and independently influencing brain growth, may be hypoxia. Previous studies in fetal sheep with an hypoxic insult at a similar stage of brain development show reductions in brain weight, SFI, and process growth.18 In this baboon model, as in the preterm infant, attempts were made to reduce hypoxic episodes, but despite this there is variation in the animals' respiratory stability, which seems to relate to the extent of gliosis and white matter volumes. The animals with lower fluxes in the fraction of inspired oxygen required to maintain appropriate blood gases had lower brain damage indices and a lower density of cortical astrocytes. These correlations did not distinguish between EnCPAP and DnCPAP. Elevated PaCO2 was also associated with an increase in white matter astrocytosis. Both elevated FIO2 and PaCO2 may reflect "sicker" animals with poor ventilatory control and lung injury. Permissive hypercapnia has become more commonly tolerated in NICU, but there are increasing concerns over a loss of cerebral autoregulation36 and increased intraventricular hemorrhage.37 These data would continue to raise concerns about the risk of hypercapnia and lung injury on brain injury. It is important to note that although these relationships are described as linear, there is probably a threshold below which there is no risk of cerebral injury for all of these physiologic measures. Unfortunately, because of the small numbers in this model, we are not able to delineate this further but may, on combination of all primate models, be able to explore this.
In addition to hypoxia and potential reduced cerebral oxygen delivery, there are many other potential contributing factors to the risk of white matter injury, gray matter gliosis, and reduced brain growth, including nutrition, infection, drugs, and animal stress and handling. Although every effort was made to provide infant baboons with adequate nutrition,14 the slow body and brain weight gain might indicate a reduced supply of nutrients to the brain, which could also have contributed to brain alterations. The DnCPAP group clearly experienced an extended time required on PPV, and this may have activated inflammatory cascades. We know from companion studies on the lungs of these animals that they have a distinct cellular bronchiolitis and cytokine levels in the bronchoalveolar lavage fluid compared with controls.16 The increased intervention procedures that are required with mechanical ventilation may also have resulted in increased stress levels with adverse effects on brain growth. Stress, at least in the form of maternal separation, is known to alter normal cerebral development.38
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
We are grateful to Vicki Winter for organizing the shipment of brains to Dr Rees' laboratory; to personnel at the Bronchopulmonary Dysplasia Resource Centre, San Antonio, TX, particularly the animal husbandry group and NICU technicians; and to Prof Hannah Kinney and Dr Rebecca Folkerth for invaluable advice on brain neuropathology.
|
FOOTNOTES |
|---|
Address correspondence to Michelle Loeliger, PhD, Department of Anatomy and Cell Biology, University of Melbourne, 3010, Victoria, Australia. E-mail: m.loeliger{at}unimelb.edu.au
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  V. Degos, G. Loron, J. Mantz, and P. Gressens Neuroprotective Strategies for the Neonatal Brain Anesth. Analg., June 1, 2008; 106(6): 1670 - 1680. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
