Brain Advance Access originally published online on February 10, 2005
Brain 2005 128(5):1070-1081; doi:10.1093/brain/awh436
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Brain damage and axonal injury in a Scottish cohort of neonatal deaths
1 Division of Pathology and 2 Division of Child Life and Health, University of Edinburgh, Edinburgh, UK
Correspondence to: Jeanne Bell BSc MD FRCPath FRSE, Professor of Neuropathology, Division of Pathology (Neuropathology), Alexander Donald Building, Western General Hospital, Edinburgh EH4 2XU, UK E-mail: Jeanne.Bell{at}ed.ac.uk
.
Received July 14, 2004. Revised November 22, 2004. Accepted January 18, 2005.
 |
Summary |
|---|
 Top Summary IntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Despite the clinical and medicolegal significance attached to perinatal asphyxia, the neuropathological basis of this condition remains obscure. There are very few studies in the literature which correlate the pathological findings in neonatal brains with detailed epidemiological data, and none which are population based. In a Scotland-wide study of neonatal deaths, 70 brains have been examined. On the basis of glial and macrophage reactions, we previously identified infants with putative antepartum brain damage in this cohort and have related these reactions to signs of birth asphyxia. The present study explores the extent of neuronal/axonal injury in these infants since this is likely to be the basis for neurological deficits in surviving infants. We have also investigated these brains for ß-amyloid precursor protein (ßAPP) positivity to determine whether this is a useful marker of neuronal injury in neonates. Neuronal eosinophilia and karyorrhexes were detected in 43% and 27% of the cohort, respectively; maximally in the subiculum and ventral pons, but often present elsewhere. White matter damage was detected in 24% of cases but without classic cystic lesions of periventricular leucomalacia. ßAPP positivity was present in neuronal soma in 52% of cases and, in axons, in 27% of cases, and was seen from as early as 25-weeks gestation. Axonal bulbs were clearly delineated by ßAPP positivity and were usually located in the cerebral white matter and internal capsule, and infrequently in the brain stem. Although white matter damage and ßAPP axonal positivity were often detected in the same cases (P = 0.034), these features also occurred independently of each other. Both neuronal karyorrhexes and white matter ßAPP positivity were significantly correlated with the features of birth asphyxia, particularly a history of seizures. Immunocytochemistry for both ßAPP and glial fibrillary acidic protein proved useful in detecting neuropathological features which escaped detection on routine examination, particularly in preterm infants. The presence together of recent and older damage in individual brains suggests that there is an ongoing neuronal response to cerebral insults. We find that ßAPP is a useful marker of white matter damage in the neonatal brain. Immunopositivity for ßAPP in these circumstances is not attributable to inflicted or accidental trauma. While birth-related trauma cannot be ruled out, hypoxia/ischaemia is a likely cause in these infants. However, the exact pathogenesis of neuronal/axonal injury in the neonatal brain remains unclear.
Key Words: neonatal death; neuropathology; amyloid precursor protein; axonal injury; antepartum brain damage
Abbreviations: ßAPP = ß amyloid precursor protein; GFAP = glial fibrillary acidic protein; H&E = haematoxylin and eosin
|
Introduction |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The origins of childhood neurological deficits are still poorly understood. Complications of pregnancy and adverse peripartum events are of uncertain predictive value. Perinatal asphyxia in the majority of cases is not followed either by neonatal encephalopathy or by cerebral palsy (Hall, 1989
; Nelson, 1989; Nelson and Leviton, 1991; Blair and Stanley, 1993; Yudkin et al., 1995; Edwards and Nelson, 1998) and there is international consensus (the International Cerebral Palsy Task Force) that encephalopathy and cerebral palsy are infrequently caused by perinatal asphyxia (Bakketeig, 1999; MacLennan, 1999). However preterm delivery is a recognized risk factor for such conditions (Murphy et al., 1995, 1997; Wood et al., 2000).
Major cohort studies of neonatal encephalopathies and related clinical conditions have shown that the associative factors are surprisingly diverse (Badawi et al., 1998a, b). However, these studies seldom include data on neuropathological findings of the deceased infants. On the other hand, reported perinatal neuropathological studies usually represent the collected experience of specialist referral centres (Gilles and Murphy, 1969; Rorke, 1982; Levene et al., 1985; Pape and Wigglesworth, 1989; Squier and Keeling, 1991; Marin-Padilla, 1996, 1997, 1999), or focus on particular age groups such as preterm infants (Skullerud and Westre, 1986; Leviton and Paneth, 1990; Paneth et al., 1990; Golden et al., 1997; Gilles et al., 1998). Only a small number of studies have investigated the clinical correlates of identified neuropathological features (Ellis et al., 1988; Mito et al., 1993; Gaffney et al., 1994). Although in recent years neuroimaging has proved to be a powerful tool for studying the extent and location of lesions in the perinatal brain (Cowan et al., 2003), neuropathological examination may be more sensitive for defining the nature of lesions. There is a complete dearth of studies designed to match detailed neuropathology with equally detailed demographic data in unselected groups of neonates.
The Scottish Perinatal Neuropathology Study was set up to investigate these issues and, in particular, to establish the prevalence of brain damage in a national cohort of perinatal deaths, including stillbirths and early neonatal deaths from 24 weeks gestation to 1 week of postnatal age, occurring during a 2 year period in the late 1990s. We recently reported that evidence of antepartum brain damage in the neonates in this cohort was more prevalent in those displaying signs of birth asphyxia than in non-asphyxiated infants (57% versus 8%), particularly in those who had suffered seizures (90%) (P < 0.005) (Becher et al., 2004). The assessment of brain damage in these neonatal deaths was based mainly on the detection of glial and microglial/macrophage responses. However, the important question of neuronal damage was not explored in our previous study.
In the present study, we further investigated this group of neonates to determine the prevalence of neuronal damage, and particularly of axonal injury, in relation to reactive changes in supporting cells and to the clinical history. One reliable marker of axonal injury, ß amyloid precursor protein (ßAPP), has rarely been investigated in the human infant brain other than in medicolegal cases (Arai et al., 1995; Baiden-Amissah et al., 1998; Geddes et al., 2001a, b; Reichard et al., 2003). It is recognized that both traumatic and hypoxic axonal injury is associated with ßAPP positivity (Geddes et al., 2001a, b). In this study, we examined 70 neonatal brains for the presence of ßAPP positivity in relation to other evidence of brain damage and to the clinical history.
|
Material and methods |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
In the Scottish Perinatal Neuropathology Study, we set out to examine the brain in all neonatal deaths (24 weeks gestation to 1 week postnatal) occurring during a two-year period in Scotland. Infants with chromosomal, cardiac and CNS abnormalities and with CNS infections were excluded. Post-mortem examination was achieved in 51% of the eligible cases (n = 137 in total, 70 brains examined). These 70 infants were divided into those who had died within 3 days of the onset of labour (n = 59) and those who had died between 3 and 7 days (n = 11). The three-day cut-off point was selected because changes such as macrophage accumulation and astrocytic hyperplasia are thought to require 3 days to become evident, from which we inferred an antepartum origin for these reactions.
Cases were co-ordinated from all 22 Scottish obstetric units through obstetric, paediatric, paediatric pathology and neuropathology colleagues. Consent for research use was obtained from the parents and the study was approved in each centre by the local research ethics committee.
Clinico-epidemiological data was gathered for each case. Up to 20 paraffin blocks were prepared from each brain according to a standard protocol and included representative samples from the cerebrum and hippocampus, basal ganglia and thalami, midbrain, pons, medulla, vermis and cerebellar hemispheres. Sections were examined after staining with haematoxylin and eosin (H&E), and with antibodies to glial fibrillary acidic protein (GFAP) (Dakocytomation Ltd; Ely, UK; diluted 1:1000, with trypsin pretreatment at 37°C for 20 min), the macrophage marker CD68 (Dako; diluted 1:200, with microwave antigen retrieval) and ßAPP (Chemicon Europe Ltd; Chandlers Ford, UK; monoclonal clone 22C11, diluted 1:100, with microwave antigen retrieval).
Sections were assessed independently by two observers (J.E.B. and B.W.) without reference initially to the clinical history or to the results of the other staining procedures. Neuronal damage was assessed in routinely stained preparations by evidence of neuronal eosinophilia and/or karyorrhexis as well as the presence of axonal swellings in the white matter. The presence or otherwise of astrocytic hyperplasia, macrophage accumulation and microglial upregulation as well as ßAPP positivity in neuronal soma and axons were scored as positive or negative for each case. Concordance between the two observers was achieved in >90% of cases and discrepant cases were resolved at a multiheaded microscope and by further review (J.K.). The localization and patterns of ßAPP immunopositivity was correlated with gestation, clinical history and evidence of glial and macrophage reactions, as well as with evidence of neuronal and white matter damage assessed on routine (H&E) staining throughout the brain.
Statistical analysis was performed using the 
2 test (Fisher's exact test where sample size was <20).
|
Results |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Clinical details
Of the 70 neonates enrolled in this study, 26 were mature infants (
37 weeks of gestation) and 44 were preterm (24–36 weeks) (Fig. 1). Thirty-eight infants were delivered by Caesarean section, performed as an elective procedure in one case and as an emergency in 37 (before the onset of labour in half of these and usually precipitated by cardiotocograph abnormalities). The mature infants lived for between 15 min and 7 days with only three surviving for >3 days. The preterm infants lived for between 5 min and 5.5 days, with only eight infants surviving for >3 days. Neonatal encephalopathy was observed in 10 mature and five preterm infants. Most infants died of the effects of prematurity, or of congenital anomalies other than those of the CNS (which were excluded from the study), or of ‘anoxia’. Fifty-three of the 70 neonates displayed one or more clinical signs of perinatal asphyxia, including 15 with seizures (10 of these infants died aged 
3 days), while 17 did not appear to be asphyxiated. Perinatal asphyxia was defined in this study by the presence of one or more of the following: (i) an Apgar score of 5.0 at 5 min; (ii) a cord or initial blood pH of <7.1; and/or (iii) grade 2/3 encephalopathy. Signs of birth asphyxia were associated with a higher prevalence of antepartum brain damage (57% versus 8%, P <0.005), but were not predictive in all cases. Oligohydramnios and meconium staining of the amniotic fluid were also associated with antepartum brain damage, but there was no concordance with placental pathology. A detailed analysis of these and other clinicoepidemiological factors, which may have contributed to neonatal death and to antepartum brain damage in this cohort, has been published previously (Becher et al., 2004).
|
Routine neuropathological findings
Assessment of neuronal damage was first undertaken in sections stained with H&E. Neuronal eosinophilia (Fig. 2A), marking recent onset hypoxic change, was observed in some or all differentiated nerve cell bodies in the cerebral cortex, basal ganglia, brain stem, particularly in the ventral pons and inferior olive, and in the cerebellar dentate nuclei and Purkinje cell layer. Karyorrhectic neuronal nuclei (Fig. 2B) were observed in a similar distribution, but additionally in the dentate gyrus of the hippocampus and in the cerebellar granular layer in occasional cases. If neuronal karyorrhexes were only present in very small numbers, they were considered indistinguishable from the normal neuronal drop-out known to occur during development and were disregarded. Even when present in large numbers, the changes affected only a subpopulation of neurons in the neocortex, but were more numerous and concentrated in the subiculum of the temporal lobe (spreading to involve the entorhinal cortex and the neighbouring sectors of the cornu ammonis) and in the ventral pons. The most severely affected cases showed loss of most Purkinje cells (Fig. 2C). Neuronal eosinophilia was identified in 43% of the cohort and karyorrhexes in 27%. Occasional infants showed cortical infarction with prominent neovascularization and perivascular fibrosis (Fig. 2D). Figures 2E–G show focal and diffuse macrophage responses in the context of neuronal loss, not all of which had been suspected on routine staining (Fig. 2F). Cortical damage also results in a conspicuous glial response (Fig. 2H). Case-by-case glial and macrophage reactions are shown in Table 1.
|
|
Evidence of white matter damage in routinely stained sections was present in 24% of cases, ranging from eosinophilic homogenization of the neuropil to areas of frank infarction, in relation to which axonal retraction bulbs (Fig. 3A) were identified in three cases and with more widespread surrounding white matter damage. Moderate numbers of karyorrhectic glial nuclei were identified within damaged white matter, accompanied by macrophage infiltration and astrocytic hyperplasia (Fig. 3B). Even in apparently normal white matter (Fig. 3C), unexpected astrocytic hyperplasia was sometimes revealed by GFAP immunostaining (Fig. 3D). In contrast, Fig. 3E shows normal white matter astrocytic immunoreactivity in this age group. No foci of cystic periventricular leucomalacia were seen.
|
Small foci of perivascular mineralization were present in the central white matter in 19% of the total cohort, more commonly in the preterm infants. One case displayed extensive mineralization in the basal ganglia and internal capsule.
In nine cases, glial nuclei in both white and grey matter were enlarged, pale and prominent (Fig. 3F), resembling the Alzheimer type II astrocytes observed in hepatic encephalopathy. Although the cytoplasm of such cells was not generally prominent, they were GFAP positive (Fig. 3G) and, in most of these cases, a reactive astrocytosis was confirmed in the white matter. Some mature neonates showed prominent changes of so-called myelination gliosis (glial cells with small darkly stained nuclei and prominent asymmetric cytoplasm, often arranged in rows) (Fig. 3H), but such cells did not prove to be GFAP reactive in contrast to juxtaposed normal astrocytes (Fig. 3I). Interpretation of such cases sometimes proved difficult on routine staining, especially in the presence of incipient white matter damage, and GFAP staining proved essential for the detection of true astrocytic hyperplasia.
Immunoreactivity for ßAPP
ßAPP immunopositivity was more prevalent in grey matter (neuronal perikarya, 53%) than in white matter (axons, 27%) (Table 1) Three patterns of ßAPP positivity were observed in the cases in this study.
ßAPP was detected in neuronal cell bodies in 52% of the study cases; this occurred in both mature and preterm infants (Table 1), varying from global expression in all differentiated neurons to particular subsets of neurons in the cortex (Fig. 4A), basal ganglia, cerebellum or brain stem. The inferior olivary nuclei and neurons in the floor of the fourth ventricle, the dentate nuclei and the Purkinje cells were particularly likely to display ßAPP positive neurons. The neurons of the immature cortex in premature babies and the germinal matrix cells and migrating foci were generally negative. Staining of the perikaryon occurred both in morphologically normal cells and in those which proved to be eosinophilic or frankly karyorrhectic on routine staining, although not all karyorrhectic neurons were positive (Fig. 4B). Within the hippocampus, the CA1 sector neurons were occasionally unstained when the rest of the cornu ammonis and the neurons of the subiculum were positive.
|
Axonal ßAPP positivity was detected in 27% of study cases (Table 1) and was seen predominantly in the periventricular central white matter of the cerebral hemisphere and the white matter of the cerebellum. Well-demarcated positive axons were particularly common in the internal capsule (Fig. 4C). Staining patterns varied from smudgy irregular patches of positivity in otherwise apparently normal white matter, to clearly demarcated single axons or bundles of axons, often with evidence of axonal beading. Axonal bulbs and varicosities were most numerous in association with areas of infarction or haemorrhage and were strongly positive for ßAPP. The brain stem showed only occasional axonal positivity, particularly in the cerebellar peduncles, in contrast with frequent positivity observed in neuronal soma in the stem.
Perivascular and white matter mineralizing foci were strongly positive for ßAPP in all the cases in which this feature had been noted in routinely stained sections (15 out of the 70 study cases; 21%) (Fig. 4D), including the case with significant mineralization in the internal capsule (Fig. 4E). These foci were particularly common in the basal ganglia and associated white matter. Mineralized foci did not stain positively with antibodies to GFAP and CD68.
Each of these patterns of ßAPP positivity was sometimes found in isolation but much more frequently in association with glial and macrophage responses (Table 1). ßAPP positivity was found occasionally in neurons in cases in which no other evidence of damage had been detected. Conversely, in five cases displaying microglial/macrophage and/or glial responses suggestive of antepartum damage, no ßAPP positivity was detected. Although faint ßAPP positivity was noted in reactive astrocytes, glia and blood vessels were generally negative for this marker.
Table 2 summarizes the prevalence of ßAPP positivity in different groups of infants in the present study. One case was not available for ßAPP staining, but of those available, 37 (52%) showed neuronal and 19 (28%) showed axonal ßAPP positivity. The prevalence of ßAPP in the 59 infants dying at 3 days was significantly related to the timing of damage, being higher in both neuronal bodies (P = 0.08) and axons (P = 0.03) in infants with antepartum damage. Even some infants with no apparent damage on routine staining showed occasional ßAPP positivity (27% in neurons and 9% in axons). In the 11 infants dying between 4 and 7 days of age, 36% (and all of the three infants born at term) showed both neuronal and axonal ßAPP positivity. In all groups, preterm infants showed a lower prevalence of ßAPP positivity in neurons and axons. Table 2 also shows the declining prevalence of ßAPP positivity in neonates with postnatal and no observable damage compared with those displaying antepartum damage. In contrast, the infants aged 3 days displayed the highest prevalence of axonal ßAPP positivity. White matter damage (24%) and ßAPP positivity (27%) were usually present together (P = 0.034).
|
Correlation of neuropathological findings with clinical features
Neuronal eosinophilia was no more common in asphyxiated infants (43%, n = 53) than in non-asphyxiated infants (41%, n = 17) (P = 0.87). However, this feature was detected more frequently in mature (62%, n = 26) than preterm (32%, n = 44) infants (P = 0.015). In contrast with neuronal eosinophilia, neuronal karyorrhexes were significantly increased in asphyxiated infants (36%) compared with non-asphyxiated infants (0%) (P < 0.01) and were also more prevalent in mature (27%) than in preterm infants (7%), although not significantly so (P = 0.08). Encephalopathic infants showed a particularly high prevalence of neuronal eosinophilia and karyorrhexes (90%). White matter damage was significantly more common in encephalopathic (90%) than in non-encephalopathic asphyxiated (29%) and non-asphyxiated (12%) neonates (P = 0.001). Infants who survived for
3 days had a higher prevalence of white matter damage than those who survived birth by only a few hours (P = 0.03). Encephalopathy was significantly associated with reactive astrocytosis (P = 0.001), with macrophage reactions (P = 0.002) and with metabolic astrocytosis (P = 0.024), but not with deposits of mineral. The minor differences in micromineralization in white matter parenchyma between asphyxiated and non-asphyxiated infants did not amount to significance.
White matter ßAPP immunopositivity was confined to the asphyxiated group (12 out of 46 cases) among infants who died at 3 days, but was not most prevalent in the encephalopathic infants. However, somatic neuronal ßAPP positivity was significantly more common in encephalopathic infants than in those infants who did not have seizures (P = 0.012) (Table 2). A minority of cases showed no evidence of ßAPP positivity despite the focal presence of reactive astrocytosis and/or macrophage infiltration of the white matter, suggesting that damaged neurons may already have been removed in these cases. Conversely in infants with postnatal brain damage, mostly in the form of fresh haemorrhage, neuronal ßAPP positivity was present in six cases in the absence of astrocytic or macrophage reactions. Overall, there was a significant correlation between white matter damage observed in routinely stained sections and axonal ßAPP positivity (P = 0.034). The prevalence of axonal ßAPP positivity following forceps or Caesarean delivery (n = 38) was 37% compared with only 16% in 32 infants delivered normally (P = 0.01).
|
Discussion |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
This paper documents the prevalence of neuronal damage and ßAPP positivity in a population based cohort of 44 preterm and 26 mature neonatal deaths ascertained in a two-year study. Based on the presence of glial and/or macrophage reactions, we have concluded that brain damage occurred before the onset of labour in 46% of the 59 infants who died at or before 3 days of age (Becher et al., 2004
). This assessment of the timing of neuropathological damage was grounded in previous observations in human infants (Norman, 1978; Low et al., 1989; Squier and Keeling, 1991; Del Bigio and Becker, 1994; Wigglesworth and Bridger, 1994; Norenberg, 1994; Squier, 2001; Kinney et al., 2002). However, our previous study failed to identify any pointers which would reliably predict the birth of an infant with antepartum brain damage. We did find an association between brain damage and oligohydramnios, meconium staining of the liquor and an abnormal cardiotocograph, but we found no evidence for an association with prenatal infection or placental inflammation.
This study focuses on neuronal damage. Certain staining and morphological changes (eosinophilia and karyorrhexis) signify irretrievable neuronal death. Most animal studies of timed cerebral insult indicate that neuronal karyorrhexis is visible within 24 h (Ferrer et al., 1994; Ferrer, 1996; Tan et al., 1998) and estimates of the time lapse between insult and karyorrhexis in the human brain are similar (Friede, 1972; Low et al., 1989; Wigglesworth and Bridger, 1994; Squier, 2001). The development of neuronal eosinophilia is thought to require at least 6 h in the rat model (Graeber et al., 2002) and possibly longer in the human infant (Norman, 1978; Low et al., 1989). From these estimates, it is clearly impossible to infer a definite antepartum origin to the neuronal changes in this cohort, in contrast with the glial and macrophage changes reported previously. With respect to eosinophilia, it is as likely that this was a postnatal event as prenatal in infants surviving for more than a few hours. However, more long-standing neuronal damage in the form of karyorrhexis may have commenced before labour in infants who died during the first day of life.
Apart from two cases of minor cortical dysplasia and one of GM1 gangliosidosis, the damage observed in this study was superimposed on otherwise normally developed brains and is similar to that reported in previous studies (Gilles and Murphy, 1969; Friede, 1972; Norman, 1978; Rorke, 1982; Roessmann and Gambetti, 1986; Skullerud and Westre, 1986; Ellis et al., 1988; Low et al., 1989; Leviton and Paneth, 1990; Paneth et al., 1990; Squier and Keeling, 1991; Mito et al., 1993; Del Bigio and Becker, 1994; Gaffney et al., 1994; Norenberg, 1994; Wigglesworth and Bridger, 1994; Marin-Padilla, 1996, 1997, 1999; Golden et al., 1997; Gilles et al., 1998; Squier, 2001; Graeber et al., 2002; Kinney et al., 2002), although the prevalence of grey matter and neuronal damage is higher in the present cases. Both necrosis and apoptosis have been implicated in perinatal neuronal loss (Edwards et al., 1997). Unequivocal signs of cell death are more easily detected in the differentiated neurons of the mature infant brain and were particularly prevalent in asphyxiated babies who suffered seizures. In very few of the present cases were all neurons affected, even within a single target area of the brain. Specific cell surface receptors may confer these differing levels of vulnerability, as well as forming the basis of perinatal patterns of neuronal involvement, as in pontosubicular necrosis (Kinney and Armstrong, 2002). The reactions of the immature brain to hypoxic/ischaemic and other injury differ from those of the adult brain. The white matter appears to be particularly vulnerable in the perinatal period and both white matter infarction and the presence of axonal swellings have been described previously (Norman, 1978; Squier, 2001), although this latter feature has not been widely recognized.
This study reports the first evaluation of ßAPP immunoreactivity in a large series of cases with perinatal brain damage. ßAPP is known to be upregulated in neuronal stress and is a recognized marker of traumatic axonal damage (Graham et al., 2000). Geddes et al. (2001a, b) have described a geographic pattern of axonal ßAPP staining in the cerebrum and in the lower brain stem, particularly in the cortico-spinal tracts, in babies with non-accidental head injury. Most of this immunopositivity, with the exception of ßAPP positive axonal bulbs, was ascribed to hypoxic rather than traumatic injury. Reichard et al. (2003) also described ßAPP immunopositive bulbs in a series of 29 medicolegal infant cases, not all of whom had sustained traumatic injury. Results in the present study show that ßAPP is a very useful marker for axonal injury in neonates as young as 25 weeks gestation. Similar findings were reported in a previous small study (Baiden-Amissah et al., 1998). Inflicted trauma was ruled out in the present cases since virtually all were under close hospital supervision throughout life. The infants displaying ßAPP positivity were delivered by a variety of methods including vaginal delivery and, in 19 cases, delivery by emergency Caesarean section, suggesting that axonal injury in these infants originates with antepartum events rather than prolonged birth-related trauma. It is apparent that ßAPP staining patterns in the developing brain should be interpreted with caution and that the presence of ßAPP positive axonal bulbs should not necessarily be ascribed to a traumatic event.
Since ßAPP positivity appears very rapidly after a neuronal insult (Sherriff et al., 1994; Baiden-Amissah et al., 1998; Graham et al., 2000), this feature is unlikely to be of use in ascribing brain damage in the neonate to the antepartum, intrapartum or postnatal period except perhaps in infants who die immediately after birth and when labour has been short. Immunodetection in the neuronal cell body implies upregulation of ßAPP expression whereas axonal positivity results from disturbed axonal transport (Sherriff et al., 1994). Uncertainties remain as to the duration and/or reversibility of these phenomena. A study of periventricular leucomalacia revealed that ßAPP immunoreactive axons which were detected around early lesions were not present in mature lesions (Arai et al., 1995). Perivascular amphophilic droplets and foci of micromineralization were consistently ßAPP positive in the present study, supporting the thesis that these features may result from breakdown of the blood brain barrier (Squier, 2001), since soluble APP is present in the circulation as well as in CSF (Mattson, 1997). ßAPP was not generally observed in immature neuronal cell bodies, unlike preterm axons. Overall, ßAPP expression was more frequently observed in mature than preterm brains.
With respect to reactive gliosis, some confusion exists in the literature between the so-called myelination gliosis of normal development and the pathological state of reactive astrocytosis (Squier, 2001; Kinney and Armstrong, 2002). The cells of ‘myelination gliosis’ are reminiscent of reactive astrocytes in that they have eccentric flares of eosinophilic cytoplasm but they are oriented in regular fashion in white matter of undamaged appearance, which is in the process of myelination. In the present study, we have found such cells to be GFAP negative and infer that they are oligodendroglia. In contrast, astrocytes which prove to be markedly reactive on GFAP immunocytochemistry may be inconspicuous on routine staining. We conclude that a meaningful assessment of gliosis in the neonatal brain requires GFAP immunocytochemistry. The significance of the apparent Alzheimer type II astrocytosis present in a number of cases is unclear, but has been remarked upon previously in perinatal brains (Kinney and Armstrong, 2002). It may represent the result of metabolic upset and, in our series, was more common in infants with damaged brains.
Correlation between ßAPP positive axonal injury and glial/macrophage reactions, although significant (P = 0.034), was lower than we had anticipated and quite a number of encephalopathic cases with reactive changes did not display white matter ßAPP positivity. This may be a sampling phenomenon, since both ßAPP reactive axons and gliosis may be focal phenomena.
With respect to pathogenesis of brain damage in the neonate, many of the pathological features noted in this study have been ascribed classically to hypoxic/ischaemic insults (Rivkin and Volpe, 1993). However, consideration should also be given to other possible mechanisms, including systemic infection and inflammation (Pape and Wigglesworth, 1989; Ellis et al., 2000; Wheater and Rennie, 2000)—possibly mediated by pro-inflammatory cytokines (Yoon et al., 1997; Duggan et al., 2001) or by free radicals (Tan et al., 1998; Kinney and Armstrong, 2000). We did not find any evidence of an association between brain damage and intrauterine infection in the Scottish study (Becher et al., 2004). The possibility exists that cytokine upregulation in the brain may be a response to neural damage, rather than its cause. Similarly, some of the observed neuronal changes including ßAPP expression may be the result rather than the cause of seizures in encephalopathic infants. However, it should be noted that these cellular changes also occurred in infants who had not had seizures. This study highlights the prevalence of recent onset neuronal damage in both mature and preterm infants often co-present with older lesions, suggesting an ongoing neuronal response to cerebral insults. Unfortunately, the pregnancy-associated clinical features that might predict the likelihood of brain damage remain elusive (Becher et al., 2004) and the exact pathogenesis of neuronal/axonal injury remains unclear in the neonatal brain.
|
Notes |
|---|
With the collaboration of the Scottish paediatric pathologists and neuropathologists Drs N. Alsanjari, D. Doyle, I. Graham, E. Gray, A. G. Howatson, S. Lang, A. M. Lutfy, J. McCullough, J. MacKenzie, K. J. McKenzie, J. McPhie, R. Nairn and N. Smith
|
Acknowledgements |
|---|
We wish to thank the midwives and medical staff in the 22 neonatal and obstetric units throughout Scotland whose enthusiasm and help in enrolling cases and collecting data made this study possible. We also wish to thank Ms Angela Penman and Dr Iain Anthony for assistance with the manuscript and illustration production. Finally, we are indebted to the families who, despite the turmoil surrounding the deaths of their infants and knowing that the results would not personally help them at their time of grief, allowed these detailed investigations to proceed. This study was funded by the Scottish Home and Health Department and by Wellbeing, and was dependent on the collaboration on the Scottish paediatric pathologists and neuropathologists named at the beginning of the paper.
|
References |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Arai Y, Deguchi K, Mizuguchi M, Takashima S. Expression of beta-amyloid precursor protein in axons of periventricular leukomalacia brains. Ped Neurol 1995; 13: 161–3.[CrossRef][Web of Science][Medline]
Badawi N, Kurinczuk JK, Keogh JM, Alessandri LM, O'Sullivan F, Burton PR, et al. Antepartum risk factors for newborn encephalopathy: the Western Australia case-control study. BMJ 1998a; 317: 1549–53.
Badawi N, Kurinczuk JK, Keogh JM, Alessandri LM, O'Sullivan F, Burton PR, et al. Intrapartum risk factors for newborn encephalopathy: the Western Australian case-control study. BMJ 1998b; 317: 1554–8.
Baiden-Amissah K, Joashi U, Blumberg R, Mehmet H, Edwards DA, Cox M. Expression of amyloid precursor protein (ß-APP) in the neonatal brain following hypoxic ischaemic injury. Neuropathol App Neurobiol 1998; 24: 346–52.
Bakketeig LS. Only a minor part of cerebral palsy cases begin in labour. BMJ 1999; 319: 1016–7.
Becher J-C, Bell JE, Keeling JW, McIntosh N, Wyatt B. The Scottish Perinatal Neuropathology Study—clinico-pathological correlation in early neonatal deaths. Arch Dis Child Fetal Neonatal Ed 2004; 89: F399–407.
Blair E, Stanley FJ. When can cerebral palsy be prevented? The generation of causal hypotheses by multivariate analysis of a case-control study. Ped Perinat Epidemiol 1993; 7: 272–301.[Medline]
Cowan F, Rutherford M, Groenendaal F, Eken P, Mercuri E, Bydder GM, et al. Origin and timing of brain lesions in term infants with neonatal encephalopathy. Lancet 2003; 361: 736–42.[CrossRef][Web of Science][Medline]
Del Bigio MR, Becker LE. Microglial aggregation in the dentate gyrus: a marker of mild hypoxic-ischaemic brain insult in human infants. Neuropathol App Neurobiol 1994; 20: 144–51.[CrossRef]
Duggan PJ, Maalouf EF, Watts TL, Sullivan MHF, Counsell SJ, Allsop J, et al. Intrauterine T-cell activation and increased proinflammatory cytokine concentrations in preterm infants with cerebral lesions. Lancet 2001; 358: 1699–700.[CrossRef][Web of Science][Medline]
Edwards AD, Yue X, Cox P, Hope PL, Azzopardi DV, Squier MV, et al. Apoptosis in the brains of infants suffering intrauterine cerebral injury. Ped Res 1997; 42: 684–9.[Web of Science][Medline]
Edwards DA, Nelson KB. Neonatal encephalopathies: time to reconsider the cause of encephalopathies. BMJ 1998; 317: 1537–8.
Ellis WG, Goetzman BW, Lindenberg JA. Neuropathologic documentation of prenatal brain damage. Am J Dis Child 1988; 142: 858–66.
Ellis M, Manandhar N, Manandhar DS, de L Costello AM. Risk factors for neonatal encephalopathy in Kathmandu, Nepal, a developing country: unmatched case-control study. BMJ 2000; 320: 1229–36.
Ferrer I. Cell death in the normal developing brain, and following ionizing radiation, methyl-azoxymethanol acetate, and hypoxia-ischaemia in the rat. Neuropathol Appl Neurobiol 1996; 22: 489–94.[Web of Science][Medline]
Ferrer I, Tortosa A, Macaya A, Sierra A, Moreno D, Munell F, et al. Evidence of nuclear DNA fragmentation following hypoxia-ischaemia in the infant rat brain, and transient forebrain ischaemia in the adult gerbil. Brain Pathol 1994; 4: 115–22.[Web of Science][Medline]
Friede RL. Ponto-subicular lesions in perinatal anoxia. Arch Pathol 1972; 94: 343–54.[Web of Science][Medline]
Gaffney G, Squier MV, Johnson A, Sellers S, Flavell V. Clinical associations of prenatal ischaemic white matter injury. Arch Dis Child 1994; 70: 101–6.
Geddes JF, Hackshaw AK, Vowles GH, Nickols CD, Whitwell HL. Neuropathology of inflicted head injury in children I. Patterns of brain damage. Brain 2001a; 124: 1290–8.
Geddes JF, Vowles GH, Hackshaw AK, Nickols CD, Scott IS, Whitwell HL. Neuropathology of inflicted head injury in children II. Microscopic brain injury in infants. Brain 2001b; 124: 1299–306.
Gilles FH, Murphy SF. Perinatal telencephalic leucoencephalopathy. J Neurol Neurosurg Psychiatr 1969; 32: 404–13.
Gilles FH, Leviton A, Jeffrey SM, Golden A, Paneth N, Rudelli RD. Groups of histopathologic abnormalities in brains of very low birthweight infants. J Neuropathol Exp Neurol 1998; 57: 1026–34.[Web of Science][Medline]
Golden JA, Gilles FH, Rudelli RD, Leviton A. Frequency of neuropathological abnormalities in very low birth weight infants. J Neuropathol Exp Neurol 1997; 56: 472–8.[Web of Science][Medline]
Graeber MB, Blakemore WF, Kreutzberg GW, Cellular pathology of the central nervous system. In: Graham DI, Lantos PL, editors. Greenfield's neuropathology. 7th ed. London: Arnold; 2002. p. 123–9.
Graham DI, McIntosh TK, Maxwell WL, Nicoll JAR. Recent advances in neurotrauma. J Neuropathol Exp Neurol 2000; 59: 641–51.[Web of Science][Medline]
Hall DMB. Birth asphyxia and cerebral palsy. BMJ 1989; 299: 279–82.
Kinney HC, Armstrong DD. Perinatal neuropathology. In: Graham DI, Lantos PL, editors. Greenfield's neuropathology. 7th ed. London: Arnold; 2002. p. 519–606.
Levene ML, Kornberg J, Williams TH. The incidence and severity of post-asphyxial encephalopathy in full-term infants. Early Hum Devel 1985; 11: 21–6.[CrossRef][Web of Science][Medline]
Leviton A, Paneth N. White matter damage in preterm newborns—an epidemiologic perspective. Early Hum Devel 1990; 24: 1–22.[CrossRef][Web of Science][Medline]
Low JA, Robertson DM, Simpson LL. Temporal relationships of neuropathologic conditions caused by perinatal asphyxia. Am J Obstet Gynecol 1989; 160: 608–14.[Web of Science][Medline]
MacLennan A. A template for defining a causal relation between acute intrapartum events and cerebral palsy: international consensus statement. BMJ 1999; 319: 1054–9.
Marin-Padilla M. Developmental neuropathology and impact of perinatal brain damage. I: Haemorrhagic lesions in neocortex. J Neuropathol Exp Neurol 1996; 55: 758–73.[Web of Science][Medline]
Marin-Padilla M. Developmental neuropathology and impact of perinatal brain damage. II: White matter lesions of the neocortex. J Neuropathol Exp Neurol 1997; 56: 219–35.[Web of Science][Medline]
Marin-Padilla M. Developmental neuropathology and impact of perinatal brain damage III: Gray matter lesions of the neocortex. J Neuropathol Exp Neurol 1999; 58: 407–29.
Mattson MP. Cellular actions of ßamyloid precursor protein and its soluble fibrillogenic derivatives. Physiol Reviews 1997; 77: 1081–132.
Mito T, Kamei A, Takashima S, Becker LE. Clinicopathological study of pontosubicular necrosis. Neuropaediatrics 1993; 24: 204–7.[Web of Science][Medline]
Murphy DJ, Sellers S, MacKenzie IZ, Yudkin PL, Johnson AM. Case-control study of antenatal and intrapartum risk factors for cerebral palsy in very preterm singleton babies. Lancet 1995; 346: 1449–54.[CrossRef][Web of Science][Medline]
Murphy DJ, Hope PL, Johnson A. Neonatal risk factors for cerebral palsy in very preterm babies: case-control study. BMJ 1997; 314: 404–8.
Nelson KB. Relationship of intrapartum and delivery room events to long-term neurologic outcome. Clin Perinatol 1989; 16: 995–1007.[Web of Science][Medline]
Nelson KB, Leviton A. How much of neonatal encephalopathy is due to birth asphyxia? Am J Dis Child 1991; 145: 1325–31.
Norenberg MD. Astrocyte responses to CNS injury. J Neuropathol Exp Neurol 1994; 53: 213–20.[Web of Science][Medline]
Norman MG. Perinatal brain damage. Perspect Ped Pathol 1978; 4: 41–92.
Paneth N, Rudelli RD, Monte W, Rodriguez E, Pinto J, Kairam R, et al. White matter necrosis in very low birth weight infants: neuropathologic and ultrasonographic findings in infants surviving six days or longer. J Ped 1990; 116: 975–84.[CrossRef][Web of Science][Medline]
Pape KE, Wigglesworth JS. Contemporary issues in foetal and neonatal medicine: perinatal bain lesions. London: Blackwell Scientific Publications; 1989.
Reichard RR, White CL, Hladik CL, Dolinak D. ß-amyloid precursor protein staining in nonhomicidal pediatric medicolegal autopsies. Neuropathol Exp Neurol 2003; 62: 237–47.[Web of Science][Medline]
Rivkin MJ, Volpe JJ. Hypoxic-ischaemic brain injury in the newborn. Semin Neurol 1993; 13: 30–9.[Web of Science][Medline]
Roessmann U, Gambetti P. Pathological reactions of astrocytes in perinatal brain injury. Immunocytochemical study. Acta Neuropath 1986; 70: 302–7.
Rorke LB. Pathology of perinatal brain injury. New York: Raven Press; 1982.
Sherriff EE, Bridges LR, Gentleman SM, Sivalogananthan S, Wilson S. Markers of axonal injury in post mortem human brain. Acta Neuropathol 1994; 88: 433–9.[Medline]
Skullerud K, Westre B. Frequency and prognostic significance of germinal matrix haemorrhage, periventricular leukomalacia, and pontosubicular necrosis in preterm neonates. Acta Neuropathol 1986; 70: 257–61.[CrossRef][Medline]
Squier MV. Acquired diseases of the nervous system. In: Keeling JW, editor. Foetal and neonatal pathology. 3rd ed. London: Springer Verlag; 2001. p. 633–58.
Squier MV, Keeling JW. The incidence of prenatal brain injury. Neuropathol App Neurobiol 1991; 17: 29–38.[CrossRef]
Tan S, Zhou F, Nielsen VG, Wang Z, Gladson CL, Parks DA. Sustained hypoxia-ischaemia results in reactive nitrogen and oxygen species production and injury in the premature fetal rabbit brain. J Neuropathol Exp Neurol 1998; 57: 544–53.[Web of Science][Medline]
Wheater M, Rennie JM. Perinatal infection is an important risk factor for cerebral palsy in very-low-birthweight infants. Dev Med Child Neurol 2000; 42: 364–7.[CrossRef][Web of Science][Medline]
Wigglesworth JS, Bridger JE. Neuropathological clues to the timing of early brain lesions. In: Lou HC, Griesen G, Larsen JF, editors. Brain lesions in the newborn. Alfred Benzon Symposium 37. Copenhagen: Munksgaard; 1994. p. 165–73.
Wood NS, Marlow N, Costeloe K, Gibson AT, Wilkinson AR. Neurologic and developmental disability after extremely preterm birth. EPICure Study Group. N Eng J Med 2000; 343: 378–84.
Yoon BH, Jun JK, Romero R, Park KH, Gomez R, Choi JH, Kim IO. Amniotic fluid inflammatory cytokines (interleukin-6, interleukin-1ß, and tumor necrosis factor-
), neonatal brain white matter lesions, and cerebral palsy. Am J Obstet Gynecol 1997; 177: 19–26.[CrossRef][Web of Science][Medline]
Yudkin PL, Johnson A, Clover LM, Murphy KW. Assessing the contribution of birth asphyxia to cerebral palsy in term singletons. Ped Perinat Epidemiol 1995; 9: 156–70.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  W. J. McCarran and M. P. Goldberg White Matter Axon Vulnerability to AMPA/Kainate Receptor-Mediated Ischemic Injury Is Developmentally Regulated J. Neurosci., April 11, 2007; 27(15): 4220 - 4229. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Fraser, L. Bennet, R. Helliwell, S. Wells, C. Williams, P. Gluckman, A. J. Gunn, and T. Inder Regional Specificity of Magnetic Resonance Imaging and Histopathology Following Cerebral Ischemia in Preterm Fetal Sheep Reproductive Sciences, February 1, 2007; 14(2): 182 - 191. [Abstract] [PDF]
|
|
|
|
 |
|
 L. Bennet, V. Roelfsema, S. George, J. M. Dean, B. S. Emerald, and A. J. Gunn The effect of cerebral hypothermia on white and grey matter injury induced by severe hypoxia in preterm fetal sheep J. Physiol., January 15, 2007; 578(2): 491 - 506. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||