| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
From the *Inserm, U676, Paris, France; Université Paris 7, Faculté de Médecine Denis Diderot, IFR02 and IFR25, Paris, France; AP HP, Hôpital Beaujon, Département d’Anesthésie Réanimation, Clichy, France; and AP HP, Hôpital Robert Debré, Service de Neurologie Pédiatrique, Paris, France.
Address correspondence and reprint requests to Pierre Gressens, MD, PhD, Inserm U676, Hôpital Robert Debré, 48 Bvd Sérrurier, 75019 Paris, France. Address e-mail to pierre.gressens{at}inserm.fr.
Abstract
Injury to the perinatal brain is a leading cause of childhood mortality and lifelong disability. Cerebral palsy and cognitive impairment are usually related to periventricular white matter damage, which is seen chiefly in babies born before 32 wk gestational age, and to corticosubcortical lesions, which occur mainly in full-term infants. Despite recent improvements in neonatal care, no effective treatment for perinatal brain lesions is available. Several interventions, such as magnesium sulfate in preterm newborns and hypothermia in term newborns, are the focus of completed or continuing clinical trials. Improved understanding of the pathophysiological mechanisms involved in perinatal brain lesions helps to identify potential targets for neuroprotective interventions, as discussed in this review.
Injury to the perinatal brain is a leading cause of death and disability in children. Despite major improvements in perinatal care, the incidence of neurological disabilities related to perinatal brain damage has not decreased significantly in Western countries over the last decades.1,2 Cerebral palsy and cognitive impairment are usually related to periventricular white matter damage (PWMD), which is seen chiefly in babies born before 32 wk gestational age, or to corticosubcortical lesions, which occur mainly in full-term infants. The formidable ethical, technical, and financial obstacles raised by research into perinatal brain damage have largely prevented the pharmaceutical industry from developing effective medications. Although there are no effective treatments for perinatal brain lesions, epidemiological and experimental data have allowed researchers to identify a number of potential targets for neuroprotective strategies. New animal models have led to the elucidation of biochemical events involved in neurodegeneration and neuroprotection.3–7 Furthermore, clinical trials of magnesium sulfate in preterm newborns or hypothermia in full-term newborns have been completed or are in progress.8,9 One important ethical difficulty is whether society and regulatory bodies wish to help the pharmaceutical industry take the risk of working in this area.
The present brief review discusses some of the pathophysiological mechanisms that underlie perinatal brain damage, with emphasis on the mechanisms that might constitute targets for neuroprotection. Although not exhaustive, this review summarizes the major currents and ideas in the field, perhaps in a way that is somewhat influenced by the authors’ conceptual bias.
PATHOPHYSIOLOGY OF PERINATAL BRAIN DAMAGE
Over the last 15 yr, the concept that perinatal brain injury is multifactorial has largely superseded the earlier belief that cardiovascular instability and hypoxia-ischemia were the sole culprits.6,7 Prenatal, perinatal, and postnatal factors that have been implicated in the pathophysiology of brain lesions associated with cerebral palsy (Fig. 1) include hypoxic-ischemic insults, maternal infection with overproduction of cytokines and other proinflammatory agents, excessive glutamate release initiating the excitotoxic cascade, oxidative stress, growth factor deficiency, specific drugs, and maternal stress.5,7–18 In addition, recent clinical studies support the existence of genetic susceptibility factors.19,20
|
Among noxious factors present in utero, some may be sufficient to cause permanent injury to the developing brain before birth, whereas others may act as predisposing or sensitizing factors ("prodamage conditions"), that increase the susceptibility to injury when a second unfavorable event occurs.8,21–24 A good understanding of this multiple-hit mechanism may be crucial to the development of effective neuroprotective strategies.
Several animal models of perinatal brain damage9 can be used to unravel the underlying molecular and cellular mechanisms and to test neuroprotective strategies. How to choose the best model for studying specific aspects of neuroprotection remains a matter of debate.9 PWMD is usually simulated by inducing infectious, inflammatory, excitotoxic, or hypoxic-ischemic insults in mammals (newborn rodents, rabbits, and cats; or fetal rats, rabbits, or sheep).9 A baboon model of preterm delivery by cesarean section25,26 is available for assessing the impact of preterm birth and various ventilation strategies on the developing white matter. The most common approach to simulating brain lesions in full-term neonates involves exposure to hypoxic-ischemic or excitotoxic insults in newborn rodents, rabbits, piglets, or dogs.9,27,28
PATHOPHYSIOLOGY OF PWMD
White-matter cells that play a key role in the pathophysiology of PWMD in animal models include preoligodendrocytes10,14,15 and microglia/macrophages.16,29,30 Preoligodendrocytes are the precursors of myelinating oligodendrocytes, which constitute a major glial population in the white matter. Preoligodendrocytes are highly vulnerable to oxidative stress, as their antioxidative defenses are limited, and to excessive glutamate exposure, as they express high levels of 
-3-amino-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) and kainate glutamatergic receptors.10,14,15 In addition, recent studies identified functional N-methyl-d-aspartate (NMDA) glutamatergic receptors on oligodendroglial processes.31,32 Brain microglia/macrophages can be activated rapidly in response to several stimuli or insults including inflammation, excess glutamate release, and hypoxia-ischemia. Activation of microglia-macrophages after excess glutamate release may involve transient expression of NMDA glutamatergic receptors by the developing white matter microglia/macrophages.16,29,30 Once activated, microglia/macrophages can release a broad array of toxic factors, including reactive oxygen and nitrogen species. Figure 2 summarizes the main cellular and molecular factors involved in the pathogenesis of glutamate-induced excitotoxic PWMD. Data from postmortem studies in humans support these experimental findings.33–35
|
Furthermore, in keeping with previous hypotheses,10 several experimental studies showed subplate neuronal cell death in models of PWMD.36,37 Subplate neurons play several important roles during brain development, including axon guidance and cortical organization. The molecular mechanisms involved in the death of subplate neurons are poorly understood. Thus, designing neuroprotective strategies that specifically target these mechanisms remains beyond reach.
Finally, data from animal models support the multiple-hit hypothesis. For example, in a mouse model of excitotoxic PWMD, exposure to proinflammatory cytokines [such as interleukin (IL)-1beta, IL-6, or tumor necrosis factor-] failed to produce brain lesions, whereas similar cytokine exposure followed by a mild excitotoxic insult induced severe PWMD.8
PATHOPHYSIOLOGY OF NEURONAL CELL DEATH
Hypoxia-ischemia (Fig. 3)38,39 induced by oxygen and glucose deprivation is followed by a severe decrease in adenosine triphosphate content, causing failure of the sodium-potassium-adenosine triphosphate pump that maintains the polarity of the neuronal membrane. Membrane depolarization induces excessive glutamate release, leading to a massive influx of sodium and calcium via the NMDA receptor. The increase in intracellular calcium activates several enzymes including phospholipases, proteases, and endonucleases, and neuronal nitric oxide (NO) synthase. Further deleterious effects occur at the reperfusion phase as a result of excessive production of superoxide (which damages the mitochondria and produces peroxynitrate by forming a complex with NO) and generation of free radicals. Reactive oxygen species cause oxidation of lipids and proteins and damage to DNA. Neuronal cell death was shown in animal models to be a combination of necrosis, apoptosis, and intermediate mechanisms. Mitochondrial damage, caspase activation, and apoptosis-inducing factor play key roles in perinatal neuronal cell death.40 Of particular interest, the delayed phase of neuronal cell death was protracted in rodents, lasting several weeks after the initial insult.41
|
Apoptosis and caspase activation have been documented in the developing brain subjected to hypoxia-ischemia. Key factors involved in apoptosis, such as caspase 3,42,43 bcl-2,44 and bax,45 are up-regulated in the immature brain compared with the adult brain.
POTENTIAL TARGETS FOR NEUROPROTECTION
Glutamate Receptors
Antagonists of the NMDA receptors for glutamate are potent neuroprotective agents in several animal models of perinatal brain lesions.46 However, NMDA receptors play key roles in successive steps of brain development, including the proliferation, migration, survival, and differentiation of neurons.47 Therefore, blocking NMDA receptors at specific neurodevelopmental stages might adversely affect brain development. Thus, in rats studied during the postnatal growth spurt, transient NMDA-receptor blockade with the potent noncompetitive antagonist MK-801 led to massive cell death by apoptosis.48 These findings constitute a strong argument against the prolonged use of potent NMDA receptor antagonists during brain development.
In contrast, drugs such as topiramate that block AMPA and kainate receptors show little neurotoxicity,49 although their effects on other steps of brain development such as synaptogenesis have not been evaluated. Topiramate is currently used as a well tolerated antiepileptic drug in adults and children older than 2-yr-of-age.50 Conventional anticonvulsants (phenytoin, phenobarbital, diazepam, clonazepam, vigabatrin, and valproate) at plasma concentrations relevant to seizure control in humans51 cause apoptotic neurodegeneration in the developing rat brain, whereas topiramate does not.49 Topiramate protected preoligodendrocytes against excitotoxic or hypoxic-ischemic death,15,52 a key event in the pathophysiology of white matter lesions in preterm infants, and protected the periventricular white matter against damage induced by an AMPA-kainate agonist in newborn mice.52
Magnesium sulfate has multiple cellular effects including blockade of NMDA receptors. It has been used for decades in pregnant women to obtain tocolysis and to treat eclampsia, with no reported adverse effects in the neonates. However, no studies have specifically addressed the consequences of magnesium exposure on brain development and neuronal apoptosis. Magnesium sulfate was neuroprotective in a model of neonatal white matter damage.53 The first multicenter controlled clinical trial where mothers at risk of delivering before 30 wk of gestation were given magnesium was completed.54 No significant perinatal side effects occurred, and neurodevelopmental benefits were noted in survivors examined at 2-yr-of-age.
Other drugs that may interfere with glutamatergic neurotransmission include riluzole, which inhibits glutamate release and interferes with the effects of proteins activated upon NMDA-receptor stimulation, and amantadine and memantine, two NMDA receptor antagonists devoid of the psychotomimetic and neurotoxic effects of phencyclidine or MK-801 when administered to adults. These drugs are neuroprotective in adults with conditions closely related to excitotoxicity. Their efficacy and safety in preterm newborns need to be determined.
Conventional Anesthetics
Histopathologic and behavioral data55,56 show that several drugs generally used for sedation exacerbate developmental neuronal cell death.57 These drugs include NMDA-receptor blockers (e.g., volatile anesthetics including nitrous oxide, ketamine, and opioid analgesics) and drugs that enhance 
-aminobutyric acidergic transmission (e.g., injectable and volatile anesthetics).57 Conventional anesthetics may exert neurotoxic effects57 in juvenile rodent models in the absence of induced brain damage. These deleterious effects in rodent models are attributed to disturbances in brain-derived neurotrophic factor (BDNF) that result in activation of apoptosis.58 These data should be viewed with caution, however. Although many studies suggest neuroprotective effects of volatile anesthetics in animal models of neonatal brain ischemia,59–62 these drugs have not yet been tested as neuroprotective in patients with PWMD. In addition, new anesthetics such as xenon may provide neuroprotection.63 As pointed out by Anand and Soriano, pain and stress in the neonatal period are associated with an increased risk for long-term adverse outcomes.64 Further experimental and clinical studies are needed to assess the relative impact of potential side effects and neuroprotective effects of anesthetics in neonates with PWMD.
Inflammation and Cytokines
Cytokines and activated microglia/macrophages may extend neuronal injury and/or sensitize the developing brain to a second insult. Therefore, interference with their effects would be expected to reduce subsequent neurological deficits. However, cytokines such as IL-1 β or IL-6 were found to exert trophic effects on neurons, at least in cell cultures.65 Similarly, activated microglia/macrophages, in addition to exerting toxic effects, can display protective properties, such as scavenging of excess glutamate through increased expression of glutamate transporters.66
Tianeptine blocked the deleterious effects of inflammatory cytokines on neonatal excitotoxic PWMD in a mouse model.67 Although its exact mechanism of action is unknown, tianeptine displays trophic properties and blocks the deleterious effects of inflammatory cytokines in several models without directly interfering with glutamate receptors.68 This last characteristic may limit potential adverse effects of tianeptine on normal brain development. Tianeptine is a well tolerated antidepressant used in human adolescents and adults. It has not been evaluated in human neonates.
IL-10 is a Th2 antiinflammatory cytokine that markedly reduces the production of proinflammatory cytokines by macrophages and downregulates the expression of activating molecules on macrophages and dendritic cells.69,70 Clinical trials of IL-10 are in progress in adults with autoimmune diseases such as psoriasis. They can be expected to provide preliminary information on the safety profile of IL-10 in humans. Similarly, the recombinant human soluble-receptor fusion protein etanercept (Enbrel®) binds to tumor necrosis factor- and is effective in a variety of inflammatory diseases (e.g., psoriasis, rheumatoid arthritis, and Crohn’s disease) in adults and children.71 Similar to tianeptine, IL-10 and etanercept block the deleterious effects of inflammatory cytokines in neonatal mice with excitotoxic brain damage.72
Several substances inhibit microglia/macrophage function and/or activation in vitro and in vivo. Chloroquine, alone or with colchicines, inhibits endocytosis, secretion, and phagocytosis by blood-derived monocytes and macrophages. Tetracyclines are broad-spectrum antibiotics that have antiinflammatory effects independent from their antimicrobial activity. Minocycline, a semisynthetic second-generation tetracycline, inhibited microglial activation and protected neurons against ischemia in adult and developing rats in several studies,73–75 although another study found that minocycline exacerbated hypoxic-ischemic cortical injury in neonatal mice.76 Similarly, chloroquine with or without colchicine, and minocycline, reduced excitotoxic microglia/macrophage activation and the resulting brain damage in newborn mice.30 Although the side effects of these drugs severely limit their use in pediatric practice, drugs that modulate microglial activation may be candidates for brain-damage prevention in high-risk neonates.
Nonspecific antiinflammatory drugs such as steroids or nonsteroidal antiinflammatory drugs (including indomethacin and ibuprofen) may reduce perinatal brain damage. These drugs are widely used to treat other conditions in premature babies. However, the possible neuroprotective effects of nonsteroidal antiinflammatory drugs remain to be characterized, both in animal models and in human studies specifically designed to test this neuroprotective hypothesis. In a recent study, cyclooxygenase-2 (Cox-2) blockade by indomethacin (a Cox-1 and Cox-2 inhibitor) or nimesulide (a specific Cox-2 inhibitor) abrogated the sensitizing effect of IL-1-β to excitotoxic brain lesions in newborn mice.77 Experimental and epidemiological studies support a protective role for antenatal steroids against PWMD,78,79 but this beneficial effect must be weighed against the adverse effects on brain development of postnatal high-dose steroids used to prevent or to treat chronic lung disease in premature infants.80–83
Oxidative Stress
Neonates, most notably preterm infants, are highly vulnerable to oxidative or nitrosative stress, as they have limited defense mechanisms against reactive oxygen and nitrogen species.84 Oxidative stress and nitrosative stress have been implicated in animal models of perinatal brain damage.85,86 Therefore, conventional methods for reducing oxidative stress and nitrosative stress may hold promise as neuroprotective treatments. Some of these methods were found effective in animal models of perinatal brain damage.12,87–89 Based on these experimental studies, allopurinol was tested in human neonates with asphyxia; no clinical benefits were documented.90 Interestingly, whereas oxidative stress and nitrosative stress were long believed to occur only during disease processes, more recent data strongly suggest that low levels of NO and reactive oxygen species are involved in normal events such as gene transduction control.91
Another way to minimize oxidative stress in neonates is to reduce exposure to pro-oxidative agents, such as oxygen and iron. Oxygen can be a major source of oxidative stress, especially during reperfusion, and several studies established that high concentrations of inhaled oxygen caused damage to the preterm brain.92 Similarly, free iron induces the formation of reactive oxygen species, and exogenous iron significantly exacerbates excitotoxic PWMD in newborn mice.93 However, this toxic effect of exogenous iron may be compensated in clinical settings by the coadministration of erythropoietin (Epo), which displays trophic properties.94 Further experimental and clinical studies are needed to clarify this point.
Preterm infants are at high risk for exposure to prolonged or repeated episodes of hypoxemia. Hypoxemic episodes can occur during the prenatal, perinatal, or postnatal period. Causes include reduced fetal perfusion, maternal hypoxia, prolonged or complicated labor, neonatal lung immaturity and diseases, cardiac instability, and recurrent apneas. Although the immature brain seems more resistant to oxygen desaturation than the adult brain, hypoxemic episodes may damage the newborn brain95 via several mechanisms: direct toxicity leading to destructive brain damage and associated functional deficits, sensitization of the developing brain to a subsequent insult, and disruption of brain development programs leading to long-term functional deficits without destructive brain lesions.
Although low oxygen levels in neonates may damage the developing brain, two other points must be considered when evaluating oxygen requirements in human preterm infants. First, in animal models, exposure to short periods of hypoxia protected against brain damage caused by subsequent insults (i.e., induced preconditioning).96 Second, massive or rapid reoxygenation after a hypoxic episode may lead to oxidative stress with deleterious effects on the immature brain.97
Prevention of Delayed Neuronal Cell Death
Proton magnetic resonance spectroscopy has identified two phases of brain energy failure after perinatal asphyxia in term newborns or corresponding animal models.98 These phases of energy failure are accompanied by phases of neuronal cell death.99 As previously mentioned, the delayed phase of neuronal cell death in rodents was protracted, lasting several weeks after the initial insult.41
Hypothermia was strongly neuroprotective in several experimental settings mimicking the brain lesions seen in term human neonates.36,100 Although hypothermia prevents secondary energy failure after asphyxia, it probably acts on multiple targets. One focus of concern is that the effects of neonatal hypothermia on normal brain development have not been studied. The neuroprotective effects of hypothermia are controversial. Factors that influence the effects of hypothermia include the target body temperature, mode of hypothermia induction (selective head cooling vs total body cooling), duration of hypothermia, and rate of rewarming. However, multicenter controlled clinical trials9,101 showed significant reductions in neurological handicap at 18-mo-of-age in infants exposed to moderate neonatal insult, without significant clinical side effects during neonatal hypothermia. Other trials are in progress.
Growth factors, such as insulin-like growth factor (IGF-1), nerve growth factor, or BDNF, which have antiapoptotic properties, can prevent asphyxic or excitotoxic neuronal death in animal models of perinatal damage.37,46,102–104 IGF-1 given to immature rats or lambs after hypoxia-ischemia significantly reduced the severity of brain damage. IGF-1 is associated with reduced caspase-3 and caspase-9 activities.102,105 The finding that IGF-1 remains effective when given up to 6 h posthypoxia suggests that IGF-1 acts on the phase of delayed neuronal loss.
In humans and several animal models, mutations in mitochondrial genes are involved in the genesis of neurological deficits, and mitochondria play a role in many apoptotic processes, most notably during the development of perinatal brain damage. However, no treatments directed at these major targets are available. Studies suggest that several trophic factors might affect mitochondrial function, through an unknown mechanism. In vitro, BDNF, but not nerve growth factor, increased rat and mouse brain mitochondrial respiratory coupling at complex 1.106 Further studies are required to confirm the presence of this mitochondrial effect in vivo and to determine its role in the neuroprotective effects of BDNF.
Neuropeptides modulate neuronal activity and may therefore modulate glutamate-induced neuronal cell death. Neuropeptides are inactivated by enzymatic proteolysis, indicating that proteolysis inhibition may hold therapeutic potential. Among the peptidases identified, neutral endopeptidase (NEP or neprilysin) is the prototypical member of the M13 family of metalloproteases and is widely distributed in various tissues. NEP is involved in the regulation and metabolism of a variety of biologically active peptides including tachykinins/ neurokinins.107 Interestingly, the NEP inhibitor racecadotril (Tiorfan®) is used in clinical practice to treat diarrhea, with a remarkably good safety profile.108 Racecadotril is rapidly and entirely metabolized to its active metabolite thiorphan. A recent study showed that systemic administration of thiorphan was neuroprotective against excitotoxic neuronal cell death in newborn mice.109 This neuroprotective effect was long-lasting and was still observed when thiorphan was administered 12 h after the insult, indicating a wide window for therapeutic intervention.
Epo, the main growth factor regulating erythropoiesis, exerts nonhematopoietic effects. Epo receptor is expressed in the central nervous system, Epo is produced by astrocytes, and Epo mRNA is expressed in response to hypoxia.110 Neuroprotective effects of Epo were documented in several models of immature brain injury, including models of neonatal stroke.111–113 No randomized trials designed to assess the long-term effects of recombinant Epo in neonates have been published. Additional work is needed to define the target population, optimal dosage, and optimal treatment duration.114,115
Caspases are effectors of apoptotic cell death and, therefore, caspase inhibitors may help to preserve neuronal function by extending the therapeutic window and providing long-term neuroprotection. Recently, a specific caspase inhibitor was shown to prevent neonatal stroke in rats.116 Currently, several inhibitors are undergoing preclinical drug development.117
Plasticity and Repair
Studies of animal models have identified a major area for therapeutic advances: although neuroprotective strategies can stop lesions from progressing, drugs with neurotrophic properties can promote lesion repair in the developing brain. Prevention and early treatment of brain lesions are the most desirable goals, but promoting postlesion plasticity is the only attainable target in many cases, given the absence of early markers for perinatal brain damage (Fig. 4).
|
Melatonin has been studied in excitotoxic models of neonatal PWMD.118 Although melatonin did not prevent the appearance of PWMD, it promoted subsequent lesion repair with axonal regrowth and/or sprouting. Behavioral studies suggest that melatonin-induced histological repair of the white matter may be accompanied with improved learning capabilities. Melatonin is safe,119 even in term newborns,120 although it has rarely been evaluated in controlled trials. Melatonin derivatives that are either undergoing development (agomelatine, Valdoxan®) or commercially available (ramelteon, Rozerem®) could be tested in controlled clinical trials.
In a similar excitotoxic model of neonatal PWMD, postlesion plasticity was induced by BDNF.121 However, BDNF is of little clinical usefulness at present, given its limited ability to cross the blood-brain barrier and its central role in multiple steps of brain development, which raises concerns about adverse effects in newborns. Potential alternatives include drugs that cross the blood-brain barrier and increase BDNF production, such as ampakines (positive allosteric modulators of AMPA receptors)122 or vasoactive intestinal peptide,123 and BDNF-expressing viral vectors that induce a long-lasting increase in BDNF production after a single intracerebral injection. In experimental studies, these agents mimicked the neuroprotective properties of BDNF.17,121,124,125 These promising data require confirmation in other preclinical models. Furthermore, evidence of functional gains should be sought.
Over the last decade, numerous studies showed that stem cells can be cultured and can undergo differentiation into many specific cell types.126 Studies suggest that stem-cell therapy may hold promise for treating degenerative brain diseases.127 Thus, the introduction of stem cells into the brain at a distance from the lesion might lead to reconstitution of the lost brain cells. Two major stem cell sources have been found suitable for generating neuronal and glial cell subpopulations, namely, multipotent embryonic stem cells (ES cells) derived from the inner mass of early-stage embryos and neural stem cells derived from early embryonic neural tissues. In vitro, ES cells differentiate spontaneously into various cell lineages. Addition of specific soluble factors to the culture medium increased specific cell populations, including neural lineages. For instance, adding vasoactive intestinal peptide to the culture medium of murine ES cells promoted neuronal differentiation of these cells.128 Similarly, in vitro, neural stem cells can differentiate into various neuronal and glial lineages, according to the culture conditions. Despite the considerable hope placed in stem cell therapy for various brain disorders, we are not aware of any experimental studies investigating the feasibility or potential benefits of grafting stem cells after perinatal brain damage.
A potential alternative to grafting exogenous stem cells is to stimulate the endogenous production of neural progenitors. A study in newborn rats showed enhanced proliferation of neural stem/progenitor cells in response to neonatal hypoxia-ischemia.129 Without pharmacologic stimulation, neurogenesis occurred in rodent models of hypoxia-ischemia; however, glial cells seemed to make up most of the new cell population.130,131 Although this approach is very promising, two conditions will have to be met before it can be further developed: evidence will have to be obtained that the newly formed neural cells survive, integrate into the existing neuronal network, and improve brain function; and drugs capable of stimulating the proliferation and/or survival of the newly produced neural cells will have to be identified.
ETHICAL ISSUES
If neonates are to receive the best possible treatment, they must be involved in clinical trials. The ethical issues raised by clinical trials in neonates are particularly complex. Problems of equipoise and informed consent raise difficult issues for desperate parents as well as volunteers.
There is a pressing need for treatments that can decrease the rate of disability related to perinatal brain damage, which is life-long. Such treatments will decrease the burden imposed by disability on the patients, their parents and other caregivers, and society. However, cohort studies of patients given medications to prevent or minimize brain lesions may detect developmental abnormalities, which may reflect the effects of residual brain lesions at specific developmental stages. Determining whether such abnormalities are related only to incomplete prevention or resolution of the brain lesions or also to the medications will prove extraordinarily challenging. Conceivably, adolescents or young adults might hold the drug companies responsible for their disabilities, even if these are less marked than would have been the case without the medications. For pharmaceutical companies, this possibility is a major deterrent to research into the field of perinatal brain damage. There is a need for an open nationwide debate about the amount of risk that can be accepted in the hope of obtaining benefits from experimental treatments.
CONCLUSIONS
Perinatal neuroprotection is a major health care priority, given the enormous burden of human suffering and financial cost caused by perinatal brain damage. Promising neuroprotective strategies are emerging. However, several factors have stunted the expansion of this field. First, lessons from studies of brain damage in adults may not apply to premature infants, so that further evaluation in this last population is needed. Neuroprotective drugs may alter the normal development of the brain, which proceeds at a fast pace during the perinatal period. Third, clinical trials in neonates raise specific ethical issues, which constitute a major obstacle to involvement of the pharmaceutical industry in research into perinatal neuroprotection. These issues need to be the focus of an open debate in our society. Finally, the limited birth rate in Western countries may constitute an additional disincentive for the pharmaceutical industry.
Several key questions remain actively debated. Whether several drugs must be used in combination to effectively block the main mechanisms underlying perinatal brain damage will have to be determined. Our ability to determine which main mechanisms operate in an individual neonate, to tailor the treatment to each patient’s needs, remains uncertain. The potential similarities and differences between the mechanisms involved in diffuse PWMD, which occurs chiefly in extremely preterm infants, and those identified in classical periventricular leukomalacia need to be identified. The feasibility of designing a strategy that blocks brain damage without disrupting normal development will have to be evaluated. Brain lesions occurring in preterm infants and full term infants share multiple risk factors and molecular mechanisms, a fact that may complicate the identification of key differences, knowledge of which would assist in developing treatment strategies that are optimal for each developmental stage. Although some of the neuroprotective strategies suggested above may improve neurological outcomes of neonates at risk for perinatal brain damage, supportive care also makes a major contribution to the final neurological outcome.22 Interventions such as mechanical ventilation, hemodynamic support, fluid management, corticosteroid therapy, blood glucose concentration control, and treatment of seizures affect brain development and brain damage; yet, the optimal management of neonates remains a matter of debate.81,132
Footnotes
Accepted for publication March 5, 2008.
Supported by two publicly funded organizations, the Inserm and the Paris 7 University; and by two nonprofit private organizations, the Fondation Motrice and the Fondation des Gueules Cassées.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  F. X. McGowan Jr and P. J. Davis Anesthetic-Related Neurotoxicity in the Developing Infant: Of Mice, Rats, Monkeys and, Possibly, Humans Anesth. Analg., June 1, 2008; 106(6): 1599 - 1602. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
