Apoptosis and necrosis in the newborn piglet brain following transient cerebral hypoxia-ischaemia

Alterations in glutathione and amino acid concentrations after hypoxia-ischemia in the immature rat brain

Hypoxic-ischemic brain injury involves an increased formation of reactive oxygen species. Key factors in the cellular protection against such agents are the GSH-associated reactions. In the present study we examined alterations in total glutathione and GSSG concentrations in mitochondria-enriched fractions and tissue homogenates from the cerebral cortex of 7-day-old rats at 0, 1, 3, 8, 14, 24 and 72 h after hypoxia-ischemia. The concentration of total glutathione was transiently decreased immediately after hypoxia-ischemia in the mitochondrial fraction, but not in the tissue, recovered, and then decreased both in mitochondrial fraction and homogenate after 14 h, reaching a minimum at 24 h after hypoxia-ischemia. The level of GSSG was approximately 4% of total glutathione and increased selectively in the mitochondrial fraction immediately after hypoxia-ischemia. The decrease in glutathione may be important in the development of cell death via impaired free radical inactivation and/or redox related changes. The effects of hypoxia-ischemia on the concentrations of selected amino acids varied. The levels of phosphoethanolamine, an amine previously reported to be released in ischemia, mirrored the changes in glutathione. GABA concentrations initially increased (0-3 h) followed by a decrease at 72 h. Glutamine levels increased, whereas glutamate and aspartate were unchanged up to 24 h after the insult. The results on total glutathione and GSSG are discussed in relation to changes in mitochondrial respiration and microtubule associated protein-2 (MAP2) which are reported on in accompanying paper [64].

Opposing effects of adenosine on the survival of glial cells exposed to chemical ischemia

Extracellular adenosine (Ado) accumulates during brain ischemia. To investigate the pathophysiological role of Ado on glial cells under ischemic conditions, we examined the effect of Ado on the survival of C6 glial cells exposed to chemical ischemia (CI). Treatment with Ado during exposure to CI showed a marked protective effect, that was mediated via intracellular transport and conversion of Ado to inosine (Ino). In contrast, Ado exacerbated CI-mediated cell death when it was added during the recovery time after exposure to CI. Ado cytotoxicity was largely mediated via intracellular transport, but conversion of Ado to Ino abolished its toxicity. Ado-induced cell death was characteristic of apoptosis, and Ado increased the expression of a pro-apoptotic product Bax but decreased that of an anti-apoptotic product Bcl-2. Ado also suppressed the induction of two stress proteins HSC70 and HSP27. Furthermore, Ado induced cytochrome c release and increased caspase-3-like activity. These results indicate the dual opposing effects of Ado on glial cell survival. Intracellular accumulation of Ado can be both cytoprotective and cytotoxic, depending on its metabolic pathway.

Magnesium sulfate treatment after transient hypoxia-ischemia in the newborn piglet does not protect against cerebral damage

Transient perinatal hypoxia-ischemia (HI) can lead to delayed cerebral damage beginning 8-24 h after resuscitation. Cerebroprotective therapies applied soon after HI may thus reduce the severity of brain injury. We have previously shown that MgSO4 administration to newborn piglets after HI fails to prevent the delayed global impairment in cerebral energy metabolism characteristic of severe brain damage. However, high extracellular concentrations of magnesium ions have been found to prevent specific excitotoxic neural cell death in vivo and in vitro. This study therefore examined the hypothesis that MgSO4 administration after HI reduces damage in some regions of the brain even though global energy metabolism is unaffected. Twelve newborn piglets were subjected to global cerebral HI by transient occlusion of both common carotid arteries and reduction of the inspired oxygen fraction to 0.12 until cerebral high-energy phosphates, measured by magnetic resonance spectroscopy, were significantly depleted. Subjects were randomly assigned to two groups of six: the first received MgSO4 (three doses, 400 mg/kg 1 h after resuscitation and 200 mg/kg at 12 and 24 h), and the second received placebo infusions. At 48 h after the start of the experiment, the piglets were killed and their brains were perfused, fixed, and embedded in paraffin wax. Five-micrometer sections were stained with hematoxylin and eosin to allow semiquantitative analysis of the severity and extent of injury to the hippocampus, cerebellum, cerebral cortex, caudate nucleus, thalamus, and striatum and the white matter tracts. There was no difference in the severity of tissue damage between the MgSO4-treated group and the placebo-treated animals in any brain region.

Correlation between caspase-3 activation and three different markers of DNA damage in neonatal cerebral hypoxia-ischemia

Caspase-3 has been identified as a key protease that, by targeting a limited number of proteins, can disrupt essential homeostatic processes and initiate an orderly disassembly of cells, including degradation of genomic DNA. We demonstrate the usefulness of an antibody specific for activated caspase-3 in a model of neonatal rat hypoxia-ischemia (Hl) and correlate the spatial and temporal activation of caspase-3 with three different markers of DNA damage and with the loss of a neuronal marker [microtubule-associated protein 2 (MAP 2)]. An oligonucleotide hairpin probe (HPP) with one base overhang in the 3' end displayed a close colocalization with caspase-3 activation at 3 h post-Hl, whereas terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) appeared later (24 h post-Hl). A monoclonal antibody against single-stranded DNA appeared to stain an entirely different population of cells, not positive for active caspase-3, HPP, or TUNEL at this time point. After 24 h of reperfusion, however, when cellular injury is extensive, all markers stained a large number of cells with a high degree of colocalization, and all markers delineated regions with loss of MAP 2. We conclude that the HPP shows the best correlation with pathological caspase-3 activation in this model.

Respiratory responses to single and episodic hypoxia during development: mechanisms of adaptation

The respiratory responses of the developmental subject to single and repeated episodes of hypoxia are distinct. During a single exposure, the fetus responds with an arrest of breathing activity, and the neonate, with excitation followed by depression (the biphasic response). Mechanisms under active consideration include chemosensory resetting, hypometabolism, prevalence of inhibitory neurotransmitter/modulator influence, and supramedullary regulation of control functions. When exposed to recurrent episodic hypoxia, neonates respond with relative hypoventilation, i.e. tolerance to a subsequent hypoxic stimulus. Whereas the investigation of processes responsible for this tolerance is at its infancy, studies using chronic hypoxia appear to be a useful guide. So far, altered interstitial neuromodulator levels and central markers of programmed neuronal death are harbingers of future research in this field. The clarification of the mechanisms involved in response to recurrent episodic hypoxia during development will be of fundamental value and may be useful for the eventual treatment and/or prevention of harmful central respiratory-related processes.

Intrauterine infection induces programmed cell death in rabbit periventricular white matter

An association between chorioamnionitis and periventricular leukomalacia has been reported in human preterm infants. However, whether this link is causal has not been convincingly established, and the underlying molecular mechanisms remain unclear. The objective of this study was to establish a reproducible model of cerebral white matter disease in preterm rabbits after intrauterine infection. Escherichia coli was inoculated into both uterine horns of laparotomized pregnant rabbits when gestation was 80% complete. The fetuses were delivered by cesarean section and killed 12, 24, or 48 h after the inoculation. Programmed cell death in the white matter was evaluated by hematoxylin-eosin-saffron staining and in situ fragmented DNA labeling (terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling). In a first group of 14 pregnant rabbits not treated with antibiotics, all fetuses delivered 48 h after inoculation were stillborn, whereas fetuses extracted 12 or 24 h after inoculation were alive. No significant cell death was detected in the live fetuses compared with the control noninfected rabbits. In a second group of five pregnant rabbits treated with ceftriaxone initiated 24 h after the inoculation and continued until cesarean section was performed 48 h after inoculation, 13 fetuses were alive, but all showed evidence of extensive programmed cell death in the white matter by hematoxylin-eosin-saffron staining and terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling. White matter damage became histologically detectable only 48 h after inoculation. Three of the 13 brains displayed periventricular white matter cysts mimicking human cystic periventricular leukomalacia. The high reproducibility of white matter damage in our model should permit further studies aimed at unraveling the molecular mechanisms of periventricular leukomalacia.

Activation of divergent neuronal cell death pathways in different target cell populations during neuroadapted sindbis virus infection of mice

Infection of adult mice with neuroadapted Sindbis virus (NSV) results in a severe encephalomyelitis accompanied by prominent hindlimb paralysis. We find that the onset of paralysis parallels morphologic changes in motor neuron cell bodies in the lumbar spinal cord and in motor neuron axons in ventral nerve roots, many of which are eventually lost over time. However, unlike NSV-induced neuronal cell death found in the brain of infected animals, the loss of motor neurons does not appear to be apoptotic, as judged by morphologic and biochemical criteria. This may be explained in part by the lack of detectable caspase-3 expression in these cells.

Novel treatments after experimental brain injury

Perinatal hypoxic-ischaemic encephalopathy(HIE) is being studied in laboratory models that allow the delayed cascade of events triggered by the energetic insult to be examined in detail. The concept of the 'excitotoxic cascade' provides a conceptual framework for thinking about the pathogenesis of HIE. Major events in the cascade triggered by hypoxia-ischaemia include overstimulation of N-methyl-D-aspartate type glutamate receptors, calcium entry into cells, activation of calcium-sensitive enzymes such as nitric oxide synthase, production of oxygen free radicals, injury to mitochondria, leading in turn to necrosis or apoptosis. New experimental approaches to salvaging brain tissue from the effects of HIE include inhibition of neuronal nitric oxide synthase, administration of neuronal growth factors, and inhibition of the caspase enzymes that execute apoptosis. Recent experimental work suggests that these approaches may be effective during a longer 'therapeutic window' after the insult, because they are acting on events that are relatively delayed. Application of modest hypothermia may allow these agents to be neuroprotective at even longer intervals after hypoxia-ischaemia.

Ventriculomegaly and reduced hippocampal volume following intrauterine growth-restriction: implications for the aetiology of schizophrenia

Structural alterations in the brains of some schizophrenic patients suggest an impairment of brain development, possibly as a result of intrauterine compromise. In this study we have tested the hypothesis that placental insufficiency during the second half of pregnancy in the guinea pig results in structural alterations similar to those seen in some schizophrenic patients. Placental insufficiency was induced in pregnant guinea pigs via uterine artery ligation at midgestation. At 60 days gestation (term: 68 days gestation) the fetal brains were prepared for quantitative histological and immunohistochemical analysis and compared with controls. Placental insufficiency resulted in growth-restricted animals with significantly larger cerebral ventricles, reduced cross-sectional area of the cerebral cortex and the striatum and reduced hippocampal volume compared with controls. There were fewer neuronal nitric oxide synthase (nNOS)-positive cells in layers 5-6 of the cingulate cortex, and in layer 1 of the frontal and temporal cortices. In contrast, there were no significant alterations in the optical density of tyrosine hydroxylase (TH), a rate-limiting enzyme in the biosynthesis of catecholamines and the dopamine transporter (DAT) in the striatum in growth-restricted animals compared with controls. These findings indicate that developmental disturbances can produce anatomical changes that resemble those found in some individuals with schizophrenia.

bcl-2 prolongs neuronal survival during hypoxia-induced apoptosis

In vivo models of hypoxic-ischemic brain injury have shown altered expression of a number of genes that are important in regulating neuronal survival. However, it is not clear as to whether hypoxia alone can alter the expression of genes regulating neuronal survival. We hypothesized that (1) hypoxia alone alters the expression of bcl-2 in neurons, (2) the severity and duration of hypoxia influence bcl-2 expression, and (3) the alteration of bcl-2 expression has an important role in regulating neuronal survival during hypoxia. Embryonic rat neocortical neurons cultured for 7-10 days were exposed to 0.1, 1, or 3% oxygen for various durations and were removed for analyses at 24-h intervals. Under all hypoxic conditions, neurons exhibited morphologic changes, as assessed by electron microscopy and Annexin V staining, consistent with apoptosis. Immunoblot and immunofluorescence analyses revealed an increase in neuronal bcl-2 protein during hypoxic exposure. Quantitative immunofluorescence analyses of bcl-2 immunostained neurons indicated that expression of bcl-2 was altered by the duration and severity of hypoxia. Attenuation of bcl-2 expression by antisense oligonucleotides decreased the proportion of surviving neurons by approximately 50% after 48 h of exposure to 0.1% oxygen. We conclude that observed increase in bcl-2, in part, plays an important role in neuronal survival during exposure to hypoxia.

Increased nitric oxide synthesis is not involved in delayed cerebral energy failure following focal hypoxic-ischemic injury to the developing brain

This study addressed the hypothesis that the delayed impairment in cerebral energy metabolism that develops 10-24 h after transient hypoxia-ischemia in the developing brain is mediated by induction of increased nitric oxide synthesis. Four groups of 14-d-old Wistar rat pups were studied. Group 1 was subjected to unilateral carotid artery ligation and hypoxia followed immediately by treatment with the nitric oxide synthase (NOS) inhibitor, Nomega-nitro-L-arginine methyl ester (L-NAME, 30 mg/kg). Group 2 underwent hypoxia-ischemia but received saline vehicle. Group 3 received L-NAME without hypoxia-ischemia, and group 4, saline vehicle alone. At defined times after insult, the expression of neuronal and inducible NOS were determined and calcium-dependent and -independent NOS activities measured. Cerebral energy metabolism was observed using 31P magnetic resonance spectroscopy. At 48 h after insult, the expression of inducible NOS increased, whereas neuronal NOS at 24 h decreased on the infarcted side. Calcium-dependent NOS activity was higher than calcium-independent NOS activity, but did not increase within 36 h after insult, and was significantly inhibited by the administration of L-NAME. However, L-NAME did not prevent delayed impairment of cerebral energy metabolism or ameliorate infarct size. These results suggest that the delayed decline in cerebral energy metabolism after hypoxia-ischemia in the 14-d-old rat brain is not mediated by increased nitric oxide synthesis.

Endothelial cell and hepatocyte deaths occur by apoptosis after ischemia-reperfusion injury in the rat liver

BACKGROUND

Ischemic injury of the liver is generally considered to result in necrosis, but it has recently been recognized that mediators of apoptosis are activated during ischemia/reperfusion. This study was designed to characterize the extent and the type of cells within the liver that undergo apoptosis at different periods of ischemia and reperfusion.

METHODS

Male Wistar rats were subjected to 30 or 60 min of normothermic ischemia. Liver sections were evaluated at the end of ischemia and at 1, 6, 24, and 72 hr after reperfusion. Apoptosis was determined by DNA fragmentation as evaluated by laddering on gel electrophoresis, in situ staining for apoptotic cells using TdT-mediated dUTP-digoxigenin nick-end labeling (TUNEL), and morphology on electron microscopy.

RESULTS

In situ staining of liver biopsy specimens using TUNEL showed significant apoptosis after reperfusion. Sinusoidal endothelial cells (SEC) showed evidence of apoptosis earlier than hepatocytes. For example, at 1 hr of reperfusion after 60 min of ischemia, 22+/-4% of the SEC stained TUNEL positive compared with 2+/-1% of the hepatocytes (P<0.001). With a longer duration of ischemia, a greater number of SEC and hepatocytes became TUNEL positive. An increase in TUNEL-positive cells was also noted with an increasing duration of reperfusion. The presence of apoptotic SEC and hepatocytes was supported by DNA laddering on gel electrophoresis and cell morphology on electron microscopy. Several Kupffer cells were seen containing apoptotic bodies but did not show evidence of apoptosis. Only rare hepatocytes showed features of necrosis after 60 min of ischemia and 6 hr of reperfusion.

CONCLUSION

These results suggest that apoptosis of endothelial cells followed by hepatocytes is an important mechanism of cell death after ischemia/reperfusion injury in the liver.

Nitric oxide-enhanced excitotoxicity-independent apoptosis of glucose-deprived neurons

Glucose deprivation has been shown to elicit neuronal death via extracellular glutamate accumulation. Here we report that immunostimulated glial expression of inducible nitric oxide synthase enhances the apoptotic death of glucose-deprived cerebellar granule cells (CGC) via the excitotoxicity-independent pathway. CGC cultures were immunostimulated by interferon-gamma (100 U/ml) and lipopolysaccharides (1 microg/ml) and 2 days later were challenged by glucose deprivation. Neither a 2-h Glucose deprivation nor a 2-day immunostimulation altered the viability of CGC. A 2-day immunostimulation, however, markedly enhanced the apoptotic death of CGC glucose-deprived for 1 h. The increased apoptotic death of glucose-deprived CGC after immunostimulation was mimicked by the nitric oxide (NO) releasing reagent 3-morpholinosydnonimine (200 microM, 30 min) and was partially prevented by the NO synthase (NOS) inhibitor N(G)-nitroarginine. The enhanced apoptotic death was not blocked by the N-methyl-D-aspartate (NMDA) receptor antagonists D-2-amino-5-phosphovalerate (APV) and dizocilpine (MK-801) or the non-NMDA receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX). Moreover, the NO-induced enhanced apoptotic death occurred without a significant increase of the concentration of glutamate in the bathing medium. Our data indicate that immunostimulated glial cells potentiate the apoptotic death of glucose-deprived CGC in part through the expression of inducible NOS but not through NMDA receptor activation. Potentiation of glucose-deprived CGC death by immunostimulated glial cells may be clinically implicated in the tendency of recurrent ischemic insults to be more severe and fatal than an initial ischemic insult.

Neuronal apoptosis in sudden infant death syndrome

Although evidence shows that victims of sudden infant death syndrome (SIDS) suffer repetitive episodes of hypoxemia, only subtle abnormalities have been found in their brains by light microscopy. The aim of the present study was to determine whether apoptosis, a form of cell death that can be triggered by hypoxemia and that leaves no scarring detectable by light microscopy, would be present in hypoxia-sensitive brain regions of SIDS victims. We looked for the presence of apoptosis with an in situ end-labeling method that detects DNA fragmentation. We studied 29 SIDS victims who were age-matched to nine control cases. We found significant neuronal apoptosis in 79% of the SIDS cases: 55% of the cases positive in the hippocampus and 96% positive in the brainstem. Whereas the distribution of apoptosis in the hippocampus was in hypoxia-sensitive subregions, the distribution in the brainstem was mostly in dorsal nuclei, including those involved with sensation in the face and position of the head (nucleus of the spinal trigeminal tract and vestibular nuclei). The control cases showed no significant apoptosis in the hippocampus and a mild degree in the brainstem in three cases. Our results indicate the occurrence of an acute insult at least several hours before death, an insult from which the infants had apparently recuperated. This suggests that SIDS victims suffered repeated apoptosis resulting in significant neuronal damage and, thus, functional loss in key brain regions. The involvement of specific nuclei in the brainstem may be linked to the fact that prone sleeping is a significant risk factor for SIDS. Enhanced neuronal death by apoptosis may thus have major implications for understanding the sequence of events leading to SIDS.

Apoptosis and programmed cell death: a role in cerebral ischemia

Hypoxic-ischemic neuronal death has long been considered to represent necrosis, but it now appears that many brain neurons undergo apoptosis after either global or focal ischemic insults. Recent studies demonstrated: 1) DNA cleavage into oligonucleosome-sized fragments demonstrated by a typical ladder pattern; 2) early endonuclease activation, as demonstrated by the presence of high molecular weight DNA fragments (300 to 50 kbp); 3) chromatin condensation and apoptotic bodies formation; 4) activation of apoptosis-associated proteins. These results may indicate that apoptosis contributes to the development of the ischemic infarct and is probably substantially distinct from ischemia-triggered excitotoxicity, which tends to produce necrosis.

Potentiation of N-methyl-D-aspartate-mediated neurotoxicity by immunostimulated murine microglia

Microglia have been shown to be immunostimulated by inflammatory cytokines and produce a number of toxic mediators. Here we report that immunostimulated microglia can synergistically enhance the N-methyl-D-aspartate (NMDA) receptor-mediated excitotoxicity in rat cerebellar granule cells (CGC) in culture. Neurotoxicity was assessed by morphological examination and by measuring the release of lactate dehydrogenase and DNA fragmentation. Cultured microglia were immunostimulated by interferon-gamma (200 U/ml) and lipopolysaccharides (10 microg/ml) and one or two days later they were used for co-culture with CGC. Co-culture of CGC with immunostimulated microglia resulted in a remarkable enhancement of the NMDA receptor-mediated death of CGC. This enhanced neurotoxicity was mimicked by the nitric oxide releaser 3-morpholinosydnonimine (SIN-1) or S-nitroso-N-acetylpenicillamine (SNAP). Superoxide dismutase and catalase, which stabilise NO by removing superoxide anion, ameliorated the potentiation of the NMDA-mediated death of CGC in co-culture with immunostimulated microglia, implying that reactions of NO with superoxide to form peroxynitrite can be implicated in the potentiated neurotoxicity. Our data indicate that immunostimulated microglia, which may involve in various neuropathologies, potentiate the NMDA receptor-mediated excitotoxicity in part through the expression of inducible nitric oxide synthase.

Brain injury in the premature infant: overview of clinical aspects, neuropathology, and pathogenesis

Brain injury in the premature infant is an extremely important problem, in part because of the large absolute number of infants affected yearly. The two principal brain lesions that underlie the neurological manifestations subsequently observed in premature infants are periventricular hemorrhagic infarction and periventricular leukomalacia. The emphases of this article are the neurology, neuropathology, and pathogenesis of these two lesions. Recent work suggests that the ultimate goal, prevention of the lesions, is potentially achievable. Periventricular hemorrhagic infarction may be preventable by prevention of germinal matrix/intraventricular hemorrhage, and periventricular leukomalacia, by detection of impaired cerebrovascular autoregulation, prevention of impaired cerebral blood flow, and interruption of the cascade to oligodendroglial cell death by such agents as free-radical scavengers.

Human oligodendroglial development: relationship to periventricular leukomalacia

Periventricular leukomalacia in the premature infant is a lesion of cerebral white matter with its greatest period of risk when white matter is immature, that is, when oligodendrocyte precursors are proliferating and differentiating, and before myelin sheaths are actively synthesized. Although the pathogenesis of perinatal cerebral white matter damage involves multiple factors, the correlation of the timing of the lesion with dominance of oligodendrocyte precursors in cerebral white matter suggests that intrinsic factors related to oligodendrocyte precursors are critical. Ischemia and infection have both been implicated as causes of perinatal white matter damage. Major mechanisms underlying oligodendrocyte injury in ischemia include glutamate toxicity, free-radical injury, and cytokine damage mediated by macrophages accompanying ischemia-induced inflammation. Factors related to a vulnerability of immature oligodendrocytes to ischemia potentially include a developmental lack of antioxidant enzymes to mediate oxidative stress. Cytokine-mediated injury to oligodendrocytes is also potentially important. A complete understanding of the role of immature white matter in the pathogenesis of periventricular leukomalacia is essential for developing strategies to prevent it.

A model of transient unilateral focal ischemia with reperfusion in the P7 neonatal rat: morphological changes indicative of apoptosis

BACKGROUND AND PURPOSE

The mechanisms leading to delayed cell death after hypoxic-ischemic injury in the developing brain remain to be elucidated. The aim of this study was to develop a model of transient focal ischemia in the neonatal rat in an attempt to create a reperfusion phase since in the filament model of reversible middle cerebral artery occlusion, size limitations precluded performing this procedure before 14 to 18 days. We then analyze whether apoptosis or necrosis occurs in this model.

METHODS

Seven-day-old Wistar rat pups (n = 96) underwent permanent left middle cerebral artery occlusion in association with 1-hour occlusion of the left common carotid artery. Evolution of the brain infarction was studied from 24 hours to 3 months on cresyl violet-stained coronal sections. Infarct volume was determined with the use of the mitochondrial stain 2,3,5-triphenyltetrazolium chloride. Neuronal death was demonstrated by the silver staining method of Gallyas et al (1980). Chromatin condensation was shown by DNA fragmentation assessed with the use of terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end-labeling (TUNEL) assay in cryostat sections and electron microscopic analysis.

RESULTS

Almost all of the animals who survived had reproducible cortical infarcts. The mean infarct volume was 31+/-7 mm3 (mean+/-SD). The ipsilateral hemisphere showed a well-delineated lesion in the frontoparietal cortex at 3-month recovery. Argyrophilic (dying) neurons were observed a few hours after reperfusion and increased with time. Cells exhibiting DNA fragmentation were shown as early as 6 hours, increased up to and peaked at 24 to 96 hours, then progressively decreased and persisted for several days, suggesting an ongoing process. Electron microscopy analysis demonstrated high condensation and clumping of chromatin beneath nuclear membrane in shrunken neurons.

CONCLUSIONS

Our study demonstrates the feasibility of performing ischemia-reperfusion in 7-day-old rats that develop progressive neuronal death with features characteristic of apoptosis. The reperfusion phase mimics events that occur during neonatal human hypoxic-ischemic encephalopathy at birth, since perinatal intensive care most often permits recirculation.

Brain injury in the premature infant--from pathogenesis to prevention

Brain injury in the premature infant is an extremely important problem, in part because of the large absolute number of infants affected yearly. The two principal brain lesions that underlie the neurological manifestations subsequently observed in premature infants are periventricular hemorrhagic infarction and periventricular leukomalacia. The emphases of this article are the neurology, neuropathology and pathogenesis of these two lesions. Recent work suggests that the ultimate goal, prevention of the lesions, is potentially achievable. Periventricular hemorrhagic infarction may be preventable by prevention of germinal matrix-intraventricular hemorrhage, and periventricular leukomalacia, by detection of impaired cerebrovascular autoregulation, prevention of impaired cerebral blood flow and interruption of the cascade to oligodendroglial cell death by such agents as free radical scavengers.