Deep Learning for Generalized EEG Seizure Detection after Hypoxia-Ischemia-Preclinical Validation.

Brain maturity and many clinical treatments such as therapeutic hypothermia (TH) can significantly influence the morphology of neonatal EEG seizures after hypoxia-ischemia (HI), and so there is a need for generalized automatic seizure identification. This study validates efficacy of advanced deep-learning pattern classifiers based on a convolutional neural network (CNN) for seizure detection after HI in fetal sheep and determines the effects of maturation and brain cooling on their accuracy. The cohorts included HI-normothermia term (n = 7), HI-hypothermia term (n = 14), sham-normothermia term (n = 5), and HI-normothermia preterm (n = 14) groups, with a total of >17,300 h of recordings. Algorithms were trained and tested using leave-one-out cross-validation and k-fold cross-validation approaches. The accuracy of the term-trained seizure detectors was consistently excellent for HI-normothermia preterm data (accuracy = 99.5%, area under curve (AUC) = 99.2%). Conversely, when the HI-normothermia preterm data were used in training, the performance on HI-normothermia term and HI-hypothermia term data fell (accuracy = 98.6%, AUC = 96.5% and accuracy = 96.9%, AUC = 89.6%, respectively). Findings suggest that HI-normothermia preterm seizures do not contain all the spectral features seen at term. Nevertheless, an average 5-fold cross-validated accuracy of 99.7% (AUC = 99.4%) was achieved from all seizure detectors. This significant advancement highlights the reliability of the proposed deep-learning algorithms in identifying clinically translatable post-HI stereotypic seizures in 256Hz recordings, regardless of maturity and with minimal impact from hypothermia.

Dysmaturation of sleep state and electroencephalographic activity after hypoxia-ischaemia in preterm fetal sheep.

Antenatal hypoxia-ischaemia (HI) in preterm fetal sheep can trigger delayed evolution of severe, cystic white matter injury (WMI), in a similar timecourse to WMI in preterm infants. We therefore examined how severe hypoxia-ischaemia affects recovery of electroencephalographic (EEG) activity. Chronically instrumented preterm fetal sheep (0.7 gestation) received 25 min of complete umbilical cord occlusion (UCO, n = 9) or sham occlusion (controls, n = 9), and recovered for 21 days. HI was associated with a shift to lower frequency EEG activity for the first 5 days with persisting loss of EEG power in the delta and theta bands, and initial loss of power in the alpha and beta bands in the first 14 days of recovery. In the final 3 days of recovery, there was a marked rhythmic shift towards higher frequency EEG activity after UCO. The UCO group spent less time in high-voltage sleep, and in the early evening (7:02 pm ± 47 min) abruptly stopped cycling between sleep states, with a shift to a high frequency state for 2 h 48 min ± 40 min, with tonic electromyographic activity. These findings demonstrate persisting EEG and sleep state dysmaturation after severe hypoxia-ischaemia. Loss of fetal or neonatal sleep state cycling in the early evening may be a useful biomarker for evolving cystic WMI.

Postischemic Infusion of Apigenin Reduces Seizure Burden in Preterm Fetal Sheep.

Seizures are common in preterm newborns and are associated with poor neurodevelopmental outcomes. Current anticonvulsants have poor efficacy, and many have been associated with upregulation of apoptosis in the developing brain. Apigenin, a natural bioactive flavonoid, is a potent inhibitor of hyaluronidase and reduces seizures in adult animal models. However, its impact on perinatal seizures is unclear. In the present study, we examined the effect of apigenin and S3, a synthetic, selective hyaluronidase inhibitor, on seizures after cerebral ischemia in preterm fetal sheep at 0.7 gestation (98-99 days, term ~147 days). Fetuses received sham ischemia (n = 9) or ischemia induced by bilateral carotid occlusion for 25 min. Immediately after ischemia, fetuses received either a continuous infusion of vehicle (0.036% dimethyl sulfoxide, n = 8) or apigenin (50 µM, n = 6). In a pilot study, we also tested infusion of S3 (2 µM, n = 3). Fetuses were monitored continuously for 72 h after ischemia. Infusion of apigenin or S3 were both associated with reduced numbers of animals with seizures, total seizure time, and mean seizure burden. S3 was also associated with a reduction in the total number of seizures over the 72 h recovery period. In animals that developed seizures, apigenin was associated with earlier cessation of seizures. However, apigenin or S3 treatment did not alter recovery of electroencephalographic power or spectral edge frequency. These data support that targeting brain hyaluronidase activity with apigenin or S3 may be an effective strategy to reduce perinatal seizures following ischemia. Further studies are required to determine their effects on neurohistological outcomes.

The neural and cardiovascular effects of exposure of gram-positive bacterial inflammation in preterm fetal sheep.

Perinatal infection or inflammation are associated with adverse neurodevelopmental effects and cardiovascular impairments in preterm infants. Most preclinical studies have examined the effects of gram-negative bacterial inflammation on the developing brain, although gram-positive bacterial infections are a major contributor to adverse outcomes. Killed Su-strain group 3 A streptococcus pyogenes (Picibanil, OK-432) is being used for pleurodesis in fetal hydrothorax/chylothorax. We therefore examined the neural and cardiovascular effects of clinically relevant intra-plural infusions of Picibanil. Chronically instrumented preterm (0.7 gestation) fetal sheep received an intra-pleural injection of low-dose (0.1 mg, n = 8) or high-dose (1 mg, n = 8) Picibanil or saline-vehicle (n = 8). Fetal brains were collected for histology one-week after injection. Picibanil exposure was associated with sustained diffuse white matter inflammation and loss of immature and mature oligodendrocytes and subcortical neurons, and associated loss of EEG power. These neural effects were not dose-dependent. Picibanil was also associated with acute changes in heart rate and attenuation of the maturational increase in mean arterial pressure. Even a single exposure to a low-dose gram-positive bacterial-mediated inflammation during the antenatal period is associated with prolonged changes in vascular and neural function.

Circadian patterns of heart rate variability in fetal sheep after hypoxia-ischaemia: A biomarker of evolving brain injury.

Hypoxia-ischaemia (HI) before birth is a key risk factor for stillbirth and severe neurodevelopmental disability in survivors, including cerebral palsy, although there are no reliable biomarkers to detect at risk fetuses that may have suffered a transient period of severe HI. We investigated time and frequency domain measures of fetal heart rate variability (FHRV) for 3 weeks after HI in preterm fetal sheep at 0.7 gestation (equivalent to preterm humans) until 0.8 gestation (equivalent to term humans). We have previously shown that this is associated with delayed development of severe white and grey matter injury, including cystic white matter injury (WMI) resembling that observed in human preterm infants. HI was associated with suppression of time and frequency domain measures of FHRV and reduced their circadian rhythmicity during the first 3 days of recovery. By contrast, circadian rhythms of multiple measures of FHRV were exaggerated over the final 2 weeks of recovery, mediated by a greater reduction in FHRV during the morning nadir, but no change in the evening peak. These data suggest that the time of day at which FHRV measurements are taken affects their diagnostic utility. We further propose that circadian changes in FHRV may be a low-cost, easily applied biomarker of antenatal HI and evolving brain injury. KEY POINTS: Hypoxia-ischaemia (HI) before birth is a key risk factor for stillbirth and probably for disability in survivors, although there are no reliable biomarkers for antenatal brain injury. In preterm fetal sheep, acute HI that is known to lead to delayed development of severe white and grey matter injury over 3 weeks, was associated with early suppression of multiple time and frequency domain measures of fetal heart rate variability (FHRV) and loss of their circadian rhythms during the first 3 days after HI. Over the final 2 weeks of recovery after HI, exaggerated circadian rhythms of frequency domain FHRV measures were observed. The morning nadirs were lower with no change in the evening peak of FHRV. Circadian changes in FHRV may be a low-cost, easily applied biomarker of antenatal HI and evolving brain injury.

Dissecting the contributions of the peripheral chemoreflex and myocardial hypoxia to fetal heart rate decelerations in near-term fetal sheep.

Brief repeated fetal hypoxaemia during labour can trigger intrapartum decelerations of the fetal heart rate (FHR) via the peripheral chemoreflex or the direct effects of myocardial hypoxia, but the relative contribution of these two mechanisms and how this balance changes with evolving fetal compromise remain unknown. In the present study, chronically instrumented near-term fetal sheep received surgical vagotomy (n = 8) or sham vagotomy (control, n = 11) to disable the peripheral chemoreflex and unmask myocardial hypoxia. One-minute complete umbilical cord occlusions (UCOs) were performed every 2.5 min for 4 h or until arterial pressure fell below 20 mmHg. Hypotension and severe acidaemia developed progressively after 65.7 ± 7.2 UCOs in control fetuses and 49.5 ± 7.8 UCOs after vagotomy. Vagotomy was associated with faster development of metabolic acidaemia and faster impairment of arterial pressure during UCOs without impairing centralization of blood flow or neurophysiological adaptation to UCOs. During the first half of the UCO series, before severe hypotension developed, vagotomy was associated with a marked increase in FHR during UCOs. After the onset of evolving severe hypotension, FHR fell faster in control fetuses during the first 20 s of UCOs, but FHR during the final 40 s of UCOs became progressively more similar between groups, with no difference in the nadir of decelerations. In conclusion, FHR decelerations were initiated and sustained by the peripheral chemoreflex at a time when fetuses were able to maintain arterial pressure. After the onset of evolving hypotension and acidaemia, the peripheral chemoreflex continued to initiate decelerations, but myocardial hypoxia became progressively more important in sustaining and deepening decelerations. KEY POINTS: Brief repeated hypoxaemia during labour can trigger fetal heart rate decelerations by either the peripheral chemoreflex or myocardial hypoxia, but how this balance changes with fetal compromise is unknown. Reflex control of fetal heart rate was disabled by vagotomy to unmask the effects of myocardial hypoxia in chronically instrumented fetal sheep. Fetuses were then subjected to repeated brief hypoxaemia consistent with the rates of uterine contractions during labour. We show that the peripheral chemoreflex controls brief decelerations in their entirety at a time when fetuses were able to maintain normal or increased arterial pressure. The peripheral chemoreflex still initiated decelerations even after the onset of evolving hypotension and acidaemia, but myocardial hypoxia made an increasing contribution to sustain and deepen decelerations.

Evolution of grey matter injury over 21 days after hypoxia-ischaemia in preterm fetal sheep.

Reduced grey matter volume in preterm infants is associated with later disability, but its time course and relationship with white matter injury are not well understood. We recently showed that moderate-severe hypoxia-ischaemia (HI) in preterm fetal sheep led to severe cystic injury 2-3 weeks later. In the same cohort we now show profound hippocampal neuronal loss from 3 days after HI. By contrast, reduction in cortical area and perimeter developed much more slowly, with maximum reduction at day 21. There was transient upregulation of cleaved caspase-3-positive apoptosis in the cortex at day 3 but no change in neuronal density or macroscopic injury of the cortex. Both microglia and astrocytes were transiently upregulated in the grey matter. EEG power was initially profoundly suppressed but partially recovered by 21 days of recovery, and final power was correlated with white matter area (p < 0.001, r2 = 0.75, F = 24.19), cortical area (p = 0.004, r2 = 0.44, F = 11.90) and hippocampi area (p = 0.049, r2 = 0.23, F = 4.58). In conclusion, the present study suggests that in preterm fetal sheep, hippocampal injury is established within a few days of acute HI, but impaired cortical growth develops slowly, in a similar time course to severe white matter injury.

Magnesium sulphate reduces tertiary gliosis but does not improve EEG recovery or white or grey matter cell survival after asphyxia in preterm fetal sheep.

Maternal magnesium sulphate (MgSO4 ) treatment is widely recommended before preterm birth for neuroprotection. However, this is controversial because there is limited evidence that MgSO4 provides long-term neuroprotection. Preterm fetal sheep (104 days gestation; term is 147 days) were assigned randomly to receive sham occlusion with saline infusion (n = 6) or i.v. infusion with MgSO4 (n = 7) or vehicle (saline, n = 6) from 24 h before hypoxia-ischaemia induced by umbilical cord occlusion until 24 h after occlusion. Sheep were killed after 21 days of recovery, for fetal brain histology. Functionally, MgSO4 did not improve long-term EEG recovery. Histologically, in the premotor cortex and striatum, MgSO4 infusion attenuated post-occlusion astrocytosis (GFAP+ ) and microgliosis but did not affect numbers of amoeboid microglia or improve neuronal survival. In the periventricular and intragyral white matter, MgSO4 was associated with fewer total (Olig-2+ ) oligodendrocytes compared with vehicle + occlusion. Numbers of mature (CC1+ ) oligodendrocytes were reduced to a similar extent in both occlusion groups compared with sham occlusion. In contrast, MgSO4 was associated with an intermediate improvement in myelin density in the intragyral and periventricular white matter tracts. In conclusion, a clinically comparable dose of MgSO4 was associated with moderate improvements in white and grey matter gliosis and myelin density but did not improve EEG maturation or neuronal or oligodendrocyte survival. KEY POINTS: Magnesium sulphate is widely recommended before preterm birth for neuroprotection; however, there is limited evidence that magnesium sulphate provides long-term neuroprotection. In preterm fetal sheep exposed to hypoxia-ischaemia (HI), MgSO4 was associated with attenuated astrocytosis and microgliosis in the premotor cortex and striatum but did not improve neuronal survival after recovery to term-equivalent age, 21 days after HI. Magnesium sulphate was associated with loss of total oligodendrocytes in the periventricular and intragyral white matter tracts, whereas mature, myelinating oligodendrocytes were reduced to a similar extent in both occlusion groups. In the same regions, MgSO4 was associated with an intermediate improvement in myelin density. Functionally, MgSO4 did not improve long-term recovery of EEG power, frequency or sleep stage cycling. A clinically comparable dose of MgSO4 was associated with moderate improvements in white and grey matter gliosis and myelin density but did not improve EEG maturation or neuronal or oligodendrocyte survival.

Alpha-adrenergic receptor activation after fetal hypoxia-ischaemia suppresses transient epileptiform activity and limits loss of oligodendrocytes and hippocampal neurons.

Exposure to hypoxic-ischaemia (HI) is consistently followed by a delayed fall in cerebral perfusion. In preterm fetal sheep this is associated with impaired cerebral oxygenation, consistent with mismatch between perfusion and metabolism. In the present study we tested the hypothesis that alpha-adrenergic inhibition after HI would improve cerebral perfusion, and so attenuate mismatch and reduce neural injury. Chronically instrumented preterm (0.7 gestation) fetal sheep received sham-HI (n = 10) or HI induced by complete umbilical cord occlusion for 25 minutes. From 15 minutes to 8 hours after HI, fetuses received either an intravenous infusion of a non-selective alpha-adrenergic antagonist, phentolamine (10 mg bolus, 10 mg/h infusion, n = 10), or saline (n = 10). Fetal brains were processed for histology 72 hours post-HI. Phentolamine infusion was associated with increased epileptiform transient activity and a greater fall in cerebral oxygenation in the early post-HI recovery phase. Histologically, phentolamine was associated with greater loss of oligodendrocytes and hippocampal neurons. In summary, contrary to our hypothesis, alpha-adrenergic inhibition increased epileptiform transient activity with an exaggerated fall in cerebral oxygenation, and increased neural injury, suggesting that alpha-adrenergic receptor activation after HI is an important endogenous neuroprotective mechanism.

Tumour necrosis factor blockade after asphyxia in foetal sheep ameliorates cystic white matter injury.

Cystic white matter injury is highly associated with severe neurodevelopmental disability and cerebral palsy in preterm infants, yet its pathogenesis remains poorly understood and there is no established treatment. In the present study, we tested the hypothesis that slowly evolving cystic white matter injury after hypoxia-ischaemia is mediated by programmed necrosis initiated by tumour necrosis factor. Tumour necrosis factor blockade was begun 3 days after hypoxia-ischaemia to target the tertiary phase of injury, when most secondary cell death is thought to be complete. Chronically instrumented preterm foetal sheep (0.7 gestation) received 25 min of hypoxia-ischaemia induced by complete umbilical cord occlusion or sham-umbilical cord occlusion (controls, n = 10), followed by intracerebroventricular infusion of the soluble TNF inhibitor, Etanercept, at 3, 8 and 13 days after umbilical cord occlusion (n = 9) or vehicle (n = 9). Foetal brains were processed for histology at 21 days after umbilical cord occlusion. Umbilical cord occlusion with vehicle was associated with a spectrum of macroscopic white matter degeneration, including white matter atrophy, ventriculomegaly and overt temporal lobe cystic white matter injury. Oligodendrocyte maturational arrest and impaired labelling of myelin proteins, characteristic of diffuse white matter injury, was observed in the parietal lobe and surrounding the cystic lesions in the temporal lobe. Etanercept markedly attenuated cystic white matter injury on the side of the intracerebroventricular infusion, with partial contralateral protection. Further, Etanercept improved oligodendrocyte maturation and labelling of myelin proteins in the temporal and parietal lobes. The present study shows that cystic white matter injury reflects late-onset tertiary cell death mediated by delayed neuroinflammation through the tumour necrosis factor pathway. Delayed tumour necrosis factor blockade markedly attenuated cystic white matter injury and restored oligodendrocyte maturation and deficits in myelin protein expression. These data suggest that delayed tumour necrosis factor blockade may represent a viable therapeutic strategy to reduce the risk of cystic and diffuse white matter injury and potentially cerebral palsy after preterm birth, with a surprisingly wide therapeutic window.

Targeting Persistent Neuroinflammation after Hypoxic-Ischemic Encephalopathy-Is Exendin-4 the Answer?

Hypoxic-ischemic encephalopathy is brain injury resulting from the loss of oxygen and blood supply around the time of birth. It is associated with a high risk of death or disability. The only approved treatment is therapeutic hypothermia. Therapeutic hypothermia has consistently been shown to significantly reduce the risk of death and disability in infants with hypoxic-ischemic encephalopathy. However, approximately 29% of infants treated with therapeutic hypothermia still develop disability. Recent preclinical and clinical studies have shown that there is still persistent neuroinflammation even after treating with therapeutic hypothermia, which may contribute to the deficits seen in infants despite treatment. This suggests that potentially targeting this persistent neuroinflammation would have an additive benefit in addition to therapeutic hypothermia. A potential additive treatment is Exendin-4, which is a glucagon-like peptide 1 receptor agonist. Preclinical data from various in vitro and in vivo disease models have shown that Exendin-4 has anti-inflammatory, mitochondrial protective, anti-apoptotic, anti-oxidative and neurotrophic effects. Although preclinical studies of the effect of Exendin-4 in perinatal hypoxic-ischemic brain injury are limited, a seminal study in neonatal mice showed that Exendin-4 had promising neuroprotective effects. Further studies on Exendin-4 neuroprotection for perinatal hypoxic-ischemic brain injury, including in large animal translational models are warranted to better understand its safety, window of opportunity and effectiveness as an adjunct with therapeutic hypothermia.

Fetal growth restriction and stillbirth: Biomarkers for identifying at risk fetuses.

Fetal growth restriction (FGR) is a major cause of stillbirth, prematurity and impaired neurodevelopment. Its etiology is multifactorial, but many cases are related to impaired placental development and dysfunction, with reduced nutrient and oxygen supply. The fetus has a remarkable ability to respond to hypoxic challenges and mounts protective adaptations to match growth to reduced nutrient availability. However, with progressive placental dysfunction, chronic hypoxia may progress to a level where fetus can no longer adapt, or there may be superimposed acute hypoxic events. Improving detection and effective monitoring of progression is critical for the management of complicated pregnancies to balance the risk of worsening fetal oxygen deprivation in utero, against the consequences of iatrogenic preterm birth. Current surveillance modalities include frequent fetal Doppler ultrasound, and fetal heart rate monitoring. However, nearly half of FGR cases are not detected in utero, and conventional surveillance does not prevent a high proportion of stillbirths. We review diagnostic challenges and limitations in current screening and monitoring practices and discuss potential ways to better identify FGR, and, critically, to identify the "tipping point" when a chronically hypoxic fetus is at risk of progressive acidosis and stillbirth.

Lack of Neuroprotection with a Single Intravenous Infusion of Human Amnion Epithelial Cells after Severe Hypoxia-Ischemia in Near-Term Fetal Sheep.

BACKGROUND: Hypoxic-ischemic encephalopathy (HIE) around the time of birth results from loss of oxygen (hypoxia) and blood supply (ischemia). Exogenous infusion of multi-potential cells, including human amnion epithelial cells (hAECs), can reduce hypoxic-ischemic (HI) brain injury. However, there are few data on treatment of severe HI in large animal paradigms at term. The aim of the current study was to determine whether infusion of hAECs early after injury may reduce brain damage after ischemia in near-term fetal sheep. METHODS: Chronically instrumented fetal sheep (0.85 gestation) received 30 min of global cerebral ischemia followed by intravenous infusion of hAECs from 2 h after the end of ischemia (ischemia-hAEC, n = 6) or saline (ischemia-vehicle, n = 7). Sham control animals received sham ischemia with vehicle infusion (sham control, n = 8). RESULTS: Ischemia was associated with significant suppression of EEG power and spectral edge frequency until the end of the experiment and a secondary rise in cortical impedance from 24 to 72 h, which were not attenuated by hAEC administration. Ischemia was associated with loss of neurons in the cortex, thalamus, striatum and hippocampus, loss of white matter oligodendrocytes and increased microglial numbers in the white matter, which were not affected by hAEC infusion. CONCLUSIONS: A single intravenous administration of hAECs did not reduce electrographic or histological brain damage after 30 min of global cerebral ischemia in near-term fetal sheep.

Cerebral Oxygenation and Metabolism After Hypoxia-Ischemia.

Perinatal hypoxia-ischemia (HI) is still a significant contributor to mortality and adverse neurodevelopmental outcomes in term and preterm infants. HI brain injury evolves over hours to days, and involves complex interactions between the endogenous protective and pathological processes. Understanding the timing of evolution of injury is vital to guide treatment. Post-HI recovery is associated with a typical neurophysiological profile, with stereotypic changes in cerebral perfusion and oxygenation. After the initial recovery, there is a delayed, prolonged reduction in cerebral perfusion related to metabolic suppression, followed by secondary deterioration with hyperperfusion and increased cerebral oxygenation, associated with altered neurovascular coupling and impaired cerebral autoregulation. These changes in cerebral perfusion are associated with the stages of evolution of injury and injury severity. Further, iatrogenic factors can also affect cerebral oxygenation during the early period of deranged metabolism, and improving clinical management may improve neuroprotection. We will review recent evidence that changes in cerebral oxygenation and metabolism after HI may be useful biomarkers of prognosis.

Is Late Prevention of Cerebral Palsy in Extremely Preterm Infants Plausible?

Preterm birth continues to be associated with neurodevelopmental problems including cerebral palsy. Cystic white matter injury (WMI) is still the major neuropathology underlying cerebral palsy, affecting 1-3% of preterm infants. Although rates have gradually fallen over time, the pathogenesis and evolution of cystic WMI are still poorly understood. Hypoxia-ischemia (HI) remains an important contributor, yet there is no established treatment to prevent injury. Clinically, serial ultrasound and magnetic resonance imaging studies typically show delayed development of cystic lesions 2-4 weeks after birth. This raises the important and unresolved question as to whether this represents slow evolution of injury occurring around the time of birth or repeated injury over many weeks after birth. There is increasing evidence that tertiary injury after HI can contribute to impairment of white and grey matter maturation. In the present review, we discuss preclinical evidence that severe, cystic WMI can evolve for many weeks after acute HI and is associated with microglia activity. This suggests the intriguing hypothesis that the tertiary phase of injury is not as subtle as often thought and that there may be a window of therapeutic opportunity for 1 to 2 weeks after hypoxic-ischemic injury to prevent delayed cystic lesions, and so, further reduce the risk of cerebral palsy after preterm birth.

Recombinant erythropoietin does not augment hypothermic white matter protection after global cerebral ischaemia in near-term fetal sheep.

Therapeutic hypothermia for hypoxic-ischaemic encephalopathy provides partial white matter protection. Recombinant erythropoietin reduces demyelination after hypoxia-ischaemia, but it is unclear whether adjunct erythropoietin treatment can further improve outcomes after therapeutic hypothermia. Term-equivalent fetal sheep received sham-ischaemia (n = 9) or cerebral ischaemia for 30 min (ischaemia-vehicle, n = 8), followed by intravenous infusion of recombinant erythropoietin (ischaemia-Epo, n = 8; 5000 IU/kg bolus dose, then 833.3 IU/kg/h), cerebral hypothermia (ischaemia-hypothermia, n = 8), or recombinant erythropoietin plus hypothermia (ischaemia-Epo-hypothermia, n = 8), from 3 to 72 h post-ischaemia. Foetal brains were harvested at 7 days after cerebral ischaemia. Ischaemia was associated with marked loss of total Olig2-positive oligodendrocytes with reduced density of myelin and linearity of the white matter tracts (P < 0.01), and microglial induction and increased caspase-3-positive apoptosis. Cerebral hypothermia improved the total number of oligodendrocytes and restored myelin basic protein (P < 0.01), whereas recombinant erythropoietin partially improved myelin basic protein density and tract linearity. Both interventions suppressed microgliosis and caspase-3 (P < 0.05). Co-treatment improved 2',3'-cyclic-nucleotide 3'-phosphodiesterase-myelin density compared to hypothermia, but had no other additive effect. These findings suggest that although hypothermia and recombinant erythropoietin independently protect white matter after severe hypoxia-ischaemia, they have partially overlapping anti-inflammatory and anti-apoptotic effects, with little additive benefit of combination therapy.

Adverse neural effects of delayed, intermittent treatment with rEPO after asphyxia in preterm fetal sheep.

KEY POINTS: We have previously shown that high-dose constant infusion of recombinant human erythropoietin (rEPO) from 30 min to 72 h after asphyxia in preterm fetal sheep reduced histological injury and improved electrophysiological recovery. This study shows that a high-dose infusion of rEPO from 6 to 72 h after asphyxia did not improve EEG recovery, oligodendrocyte and neuronal survival at 1 week post-asphyxia. Of concern, intermittent rEPO boluses started 6 h after asphyxia were associated with impaired EEG recovery and bilateral cystic injury of temporal lobe intragyral white matter. Intermittent boluses of rEPO were associated with significantly increased cerebral vascular resistance and hypoperfusion, particularly after the first dose, but did not affect seizures, suggesting mismatch between perfusion and brain activity. ABSTRACT: Recombinant human erythropoietin (rEPO) is a promising treatment for hypoxic-ischaemic brain injury. Disappointingly, a large randomized controlled trial in preterm infants found that prophylactic, repeated high-dose rEPO boluses started within 24 h of birth did not improve neurodevelopmental outcomes. We examined whether initiation of a continuous infusion of rEPO at the end of the latent phase after hypoxic-ischaemia (HI) might improve outcomes compared with intermittent bolus injections. Chronically instrumented preterm (0.7 gestation) fetal sheep received sham asphyxia or asphyxia induced by complete umbilical cord occlusion for 25 min. Six hours after asphyxia, fetuses received either a continuous infusion of rEPO (loading dose 2000 IU, infusion at 520 IU/h) from 6 to 72 h post-asphyxia or intravenous saline or 5000 IU rEPO, with repeated doses every 48 h for 5 days. Continuous infusion of rEPO did not improve EEG recovery, oligodendrocyte and neuronal survival at 1 week post-asphyxia. By contrast, intermittent rEPO boluses were associated with impaired EEG recovery and bilateral cystic injury of temporal lobe intragyral white matter in 6/8 fetuses. These studies demonstrate for the first time that initiation of intermittent rEPO boluses 6 h after HI, at a dose comparable with recent clinical trials, exacerbated neural injury. These data reinforce the importance of early initiation of many potential neuroprotective therapies.

Preventing Brain Injury in the Preterm Infant-Current Controversies and Potential Therapies.

Preterm birth is associated with a high risk of morbidity and mortality including brain damage and cerebral palsy. The development of brain injury in the preterm infant may be influenced by many factors including perinatal asphyxia, infection/inflammation, chronic hypoxia and exposure to treatments such as mechanical ventilation and corticosteroids. There are currently very limited treatment options available. In clinical trials, magnesium sulfate has been associated with a small, significant reduction in the risk of cerebral palsy and gross motor dysfunction in early childhood but no effect on the combined outcome of death or disability, and longer-term follow up to date has not shown improved neurological outcomes in school-age children. Recombinant erythropoietin has shown neuroprotective potential in preclinical studies but two large randomized trials, in extremely preterm infants, of treatment started within 24 or 48 h of birth showed no effect on the risk of severe neurodevelopmental impairment or death at 2 years of age. Preclinical studies have highlighted a number of promising neuroprotective treatments, such as therapeutic hypothermia, melatonin, human amnion epithelial cells, umbilical cord blood and vitamin D supplementation, which may be useful at reducing brain damage in preterm infants. Moreover, refinements of clinical care of preterm infants have the potential to influence later neurological outcomes, including the administration of antenatal and postnatal corticosteroids and more accurate identification and targeted treatment of seizures.