Can Adverse Neonatal Experiences
Alter Brain Development and Subsequent
Behavior?
K. J. S. Anand
Frank M. Scalzo
Department of Pediatrics, University of
Arkansas for Medical Sciences, and Pain Neurobiology
Laboratory,
Arkansas Children's Hospital Research
Institute, Little Rock, Ark., USA.
Keywords
Pain · N-Methyl-D-aspartate ·
Excitotoxicity · Apoptosis
Abstract
Self-destructive behavior in current society promotes a
search for psychobiological factors underlying this epidemic.
Perinatal brain plasticity increases the vulnerability to
early adverse experiences, thus leading to abnormal
development and behavior. Although several epidemiological
investigations have correlated perinatal and neonatal
complications with abnormal adult behavior, our understanding
of the underlying mechanisms remains rudimentary. Models of
early experience, such as repetitive pain, sepsis, or
maternal separation in rodents and other species have noted
multiple alterations in the adult brain, correlated with
specific behavioral phenotypes depending on the timing and
nature of the insult. The mechanisms mediating such changes
in the neonatal brain have remained largely unexplored. We
propose that lack of N-methyl-D-aspartate (NMDA) receptor
activity from maternal separation and sensory isolation leads
to increased apoptosis in multiple areas of the immature
brain. On the other hand, exposure to repetitive pain may
cause excessive NMDA/excitatory amino acid activation
resulting in excitotoxic damage to developing neurons. These
changes promote two distinct behavioral phenotypes
characterized by increased anxiety, altered pain sensitivity,
stress disorders, hyperactivity/attention deficit disorder,
leading to impaired social skills and patterns of
self-destructive behavior. The clinical important of these
mechanisms lies in the prevention of early insults, effective
treatment of neonatal pain and stress, and perhaps the
discovery of novel therapeutic approaches that limit neuronal
excitotoxicity or apoptosis. Copyright 2000 S. Karger
AG, Basel
[CIRP note: Apoptosis is defined by
Stedman's Medical Dictionary (1995) as "simple
deletion of scattered cells by fragmentation into membrane
bound particles which are phagocytosed by other cells;
believed to be due to programmed cell death."]
Aberrant behavior, mental illness, and drug
abuse have reached epidemic proportions amongst adolescents
and adults today. The high prevalence of such
psychopathologies imposes a signficant burden of direct and
indirect health care costs in developed countries [1, 2]. Self-destructive behaviors (e.g.,
suicide, drug abuse) are also associated with social
disruption and the propagation of sexually transmitted
diseases, leading to extensive societal and emotional
consequences [3].
While the causes of such behavior are no
doubt complex, adverse experiences around birth (many of
which are preventable) may change the course of brain
development and predispose individuals to abnormal behavior.
Indeed several investigations have correlated perinatal and
neonatal complications with these adult psychopathologies
Abnormal conditions around birth have been associated with
behavioral and emotional problems during childhood [4-8], major psychoses such as schizophrenia
[9], anxiety or depression, suicides in
adolescence [13] and adulthood [14, 15], altered responses to pain [16-18], or intractable pain states [19].
Much progress has been made in understanding
the neurobiological basis for altered adult behavior
resulting from experimental manipulations in the neonatal or
perinatal period [20, 21]. These elegant
studies have mainly focused on neuroanatomical, hormonal, and
neurochemical alterations in the adult brain casued by models
of early experience, such as repetitive pain, sepsis, or
maternal separation. However, changes occurring in the
neonatal brain as a result of these experimental
manipulations have been largely unexplored. As a result,
little is known about the pathogenesis or underlying
mechanisms that lead to the neurobiological or behavioral
changes noted in adult animals. We propose novel hypotheses
that may help to investigate the mechanisms leading to
long-term changes in adult rat behavior, resulting from
modification in brain structure and function during the
neonatal period.
Long-Term Behavioral Sequelae Related to Perinatal
Complications
Based on a cross-sectional case-control
study comparing 412 suicide victims with 2,901 matched
controls, Jacobson et al. [14] related
adult self-destructive behavior with the traumatic
experiences around the birth of these individuals. They
suggested that imprinting at birth may predispose individuals
to certain patterns of behavior that remain masked throughout
most of adult life, but may be triggered during conditions of
extreme stress [14]. In subsequent
studies, these authors controlled for genetic, socioeconomic
and demographic factors by using biological siblings as
controls. For suicides committed by violent means (e.g., by
using a firearm, jumping from a height or in front of a
train, hanging and strangulation), the significant risk
factors were those perinatal events that were likely to cause
pain in the newborn [15]. A composite
score derived from factors such as forceps or vacuum
delivery, presentation other than vertex, resuscitation and
other neonatal complications was significantly correlated
with adult suicide in male infants. [15].
Multiple risk factors markedly increased the likelihood of
the male off-spring subsequently committing violent suicide
(relative risk [RR] confidence intervals [CI] 1.8-13.0).
Maternal opioid therapy during delivery reduced the risk for
subsequent suicides (RR 0.26, 95% CI 0.09-0.69), perhaps by
making the infants less sensitive to the painful
interventions at birth [15]. Another study
[13] comparing 52 adolescents who
committed suicide and 104 matched controls found that lack of
antenatal care, chronic maternal disease during pregnancy,
and neonatal respiratory compromise increased the risk for
adolescent suicide (RR 3.42, 95% CI 1.87-5.30).
Interestingly, sedatives given in the
absence of painful interventions were noted to increase the
risk for subsequent drug abuse. Administration of multiple
doses of opiates, barbiturates and nitrous oxide to mothers
during delivery were found to increase the occurrence of
subsequent opiate (RR 4.7, 95% CI 1.8-12.0, p = 0.002) or
amphetamine (RR 5.6, 95% CI 1.6-16.9, p = 0.005) addiction in
the offspring as compared to when no drug was given [22, 23]. Obstetric factors had greater
influence on future drug addiction than socioeconomic factors
[24,25] Neonatal hypoxia was not a
significant risk factor in any of these studies [13-15, 22-25] suggesting
that such long-term effects were not associated with
extensive brain damage. Conversely, maternal smoking during
pregnancy showed a dose-response relationship with persistent
criminal behavior in male offspring, after controlling for
demographic, parental, and perinatal risk factors [26].
Full-term neonates who experienced stressful
deliveries also showed increased behavioral responses to heel
lancing and increased cortisol responses to vaccination at
4-6 months age [27]. Gunnar et al. [28, 29] reported that infants born following
obstetric complications showed altered responses to handling
and an elevated adrenocortical response to a second heelstick
performed 24 h later. Together these data suggest that
prenatal or perinatal stress may enhance subsequent responses
to stressful or painful events, and perhaps alter adult
behavior.
Long-Term Neurobehavioral Effects Related to Neonatal
Interventions
Early clinical interventions in human
neonates have multiple long-term effects [30,31] that can be briefly summarized here.
Ex-premature infants often have more educational, behavioral
and emotional difficulties during school age [32, 33] and adolescence [7,8,34] as compared to their ex-full term
peers. There is not evidence relating these long-term
outcomes to neonatal pain or stress, although such an
association may be supported by preliminary findings from the
clinical studies reviewed below.
Grunau et al. [38]
reported that 18-month-old ex-preterm neonates were less
sensitive to pain as compared with ex-full term infants. The
greater the number of procedures they had experienced as
neonates, the less responsive they were to pain during
infancy [38]. Altered behavior continued
at older ages (4-5 years), with increased somatization in the
ex-preterm infants [16], and greater
affective responses to depicted medical pain at the age of
8-10 years [17]. Thus exposure to painful
stimuli may alter the responses to subsequent stimuli,
implying that the clinical benefits of optimal pain
management may persist well beyond the duration of
therapeutic effects.
Little is known about the effects of
full-term pain in neonates. Circumcision
seemed to disrupt their post-natal adaptation [39] and treatment of circumcision pain with a
topical anesthetic decreased their responses to vaccination
pain at 4-6 months of age [18].
From other experimental studies, it is now evident that very
young infants can readily learn tasks and retain their
responses over a period of several days [40]. Together, these studies provide evidence
for classical associative learning in both newborn animals
and humans, and promote the possibility that memories for
pain may be recorded biologically [41],
but are not accessible to conscious recall [40]
Some aspects of Normal Neonatal Development
By themselves, the clinical correlations
reviewed above are not sufficient to postulate the long-term
effects of early experience. Corroborative data from animal
experiments and mechanistic explanations of how abnormal
sensory inputs in the neonatal period may lead to long-term
changes in brain structure and behavior are necessary to
substantiate these clinical observations. The following
sections describe the impact of abnormal simulation in the
newborn period within the context of maternal-infant
interactions from human and animal experiments. We propose
mechanistic hypotheses based on overstimulation or
understimulation of newborns that may provide a plausible
rationale for some of the clinical observations noted
above.
Figure 1 outlines some normal processes that
regulate early brain development and abnormal sensory
experiences that can alter the cellular mechanisms, the
organization of stress-responsive systems, and subsequent
behavior. The mechanisms for these phenomena can only be
examined within the framework of what now constitutes normal
stimulation for the newborn infant (associated with factors
such as maternal-infant interaction [42],
neonatal painful stimulation [43], brain
growth and plasticity).
Maternal-Infant Interaction
Studies of
newborn behavior have demonstrated a dynamic interactive
process of the neonatal with its mother and environment [44-46]. In this setting, both maternal and
infants behaviors play pivotal roles in producing
mother-infant bonding and providing optimal stimulation for
development of the neonate's brain. It was proposed that the
mother-infant relationship represents the prototype for
relationship with the self and all other social relationships
[47], thus highlighting its importance in
subsequent behavior.
Maternal-infant interactions were examined
by Bowlby [48-50], who then proposed the
general concepts of attachment theory. He described the
response to maternal separation in two phases, a 'protest'
phase representing highly activated attachment behavior, and
a 'despair' phase to prepare the infant for passive survival
through a combination of energy conservation and withdrawal
from danger [48]. Hofer [42] discovered the maternal regulators of
infant physiology and behavior, and classified them into
three categories: (a) thermal-metabolic, (b)
nutrient-interoceptive, and (c) behavioral-sensorimotor.
These regulated the infant's activity level, sucking
behavior, oxygen consumption, sleep-wake states, circadian
rhythms, as well as hormonal, cardiovascular, immune and
neuroendocrine responses. From this perspective, infantile
responses to maternal separation were viewed as resulting
from a single mechanisms, that is, a release from maternal
regulation. [42].
Kuhn and Schanberg [51]
noted that sensorimotor stimulation of the infant rat altered
the regulation of growth hormone secretion, and Field [52] reported the beneficial effects of
tactile-kinesthetic stimulation on weight gain and
development of preterm neonates. Recently these principles
have helped to formulate 'kangaroo care' (placement of
preterm neonates between their mother's breasts), which may
provide additional physiological and neurodevelopmental
benefits in critically ill neonates [53,
54]. Several lines of evidence underscore the importance
of maternal care-giving practices on neonatal physiology and
brain development, with similar results reported in a variety
of species [42]. The neonatal brain
appears to be well equipped for the transmission and
integration of all sensory stimuli (auditory, visual, and
tactile) required for the developmental tasks in this time
period.
The long-term effects of physical handling
or other enrichment paradigms in neonatal animals further
substantiate the importance of early childhood experiences.
For example, Pieretti et al. [55] found
that newborn mice removed from the dam for 10 min each day
developed significantly higher latencies for the tail flick
and hot plate tests at P25 and P45, and these effects were
inhibited by naloxone pretreatment. The vast literature on
the long-term effects of postnatal handling [21, 56-58] suggests highly
specific effects on adult behavior and cognitive ability.
Behavioral and Cognitive Activity of
Human Newborns
Within hours
after birth, human infants show significantly increased
preferences for their mothers voice [59],
smell [60], and facial features [61] over those of a stranger. Term and
preterm neonates actively regulate their behavioral states to
maintain an optimum level of stimulation, and unstressed
infants can smoothly transition from one behavioral state to
another [62, 63]. Even very premature
neonates actively perceive, learn, and organize information
[64], while constantly striving to control
their sensory input [62, 65]. Preterm
infants will mount a coordinated response to a painful
stimulus [66, 67] associated with changes
in their arousal, behavior, and physiology [68, 69]. Als et al. [70]
have noted that preterm infants actively approach and favor
experiences that are developmentally supportive and actively
avoid experiences that are developmentally disruptive. These
behaviors support conservation of energy, sleep-wake cycles
and the achievement of developmental milestones. Improved
clinical and neuromaturational outcomes have resulted from
developmentally supportive nursing care [36,
71] and 'kangaroo care' in preterm infants [54].
Increased Sensitivity to Pain in
Neonates
The traditional view
that neonates were relatively insensitive to pain was refuted
more than a decade ago. [66, 72]. A
comparative study of human infants and rat pups reported that
withdrawal reflex thresholds increased during gestation and
with postnatal age, suggesting lower pain thresholds at
earlier developmental stages [73].
Multiple other studies have demonstrated that the intensity
and duration of pain responses were developmentally regulated
[74-77]. Teng and Abbott [75] found that thresholds for
formalin-induced pain increased 2.4-fold from 3 to 15 days of
age, and 11-fold from the newborn to the adult rat. Thermal
stimulation using withdrawal latencies to a hot plate [76] or tail flick [77]
also reported lower pain thresholds in the younger rats.
Mechanisms underlying increased pain sensitivity in neonatal
rat pups are summarized in table 1 and described elsewhere
[78-81]. The 2 weeks after birth in new
born rat pups and the last trimester of human gestation
represent critical windows for the organization of spinal
cord mechanisms [78, 79]. During these
critical periods, the supraspinal transmission of pain
impulses at is facilitated, whereas modification of these
impulses is rudimentary. [82-84].
For many years, cortical activity was not
thought to be necessary for pain perception [85]. Studies using positron emission
tomography challenged these concepts by showing activation of
the anterior cingulate cortex, primary somatosensory cortex,
and the prefrontal cortex [86, 87].
Multiple lines of structural and functional evidence now
support the role of cortical mechanisms for pain perception
in adult humans and animals [88]. Neonatal
positron emission tomography scans also indentified
somatosensory areas as the most active sites in the
developing brain [89]. By 24-26 weeks of
gestation, putative nociceptive fibers from the
ventroposterior thalamus have fully penetrated the primary
somatosensory cortex, providing the final anatomical link for
the developing pain/tactile system [90].
Cellular organization of the somatosensory cortex and other
supraspinal areas during early development is largely
determined by the initiation of incoming afferent activity
and activity-dependent changes in gene expression.
Mechanisms of Neonatal Brain
Plasticity
Parallel
experiments in the somatosensor and visual cortex suggest
that activity-dependent effects on gene expression underlie
the establishment of cortical maps during development [91]. Current data suggest complex
interactions between early neonatal experience and the gene
products controlling cellular and neurotransmitter
development, which may lead to apoptosis in development in
developing cortical neurons. For example, insulin-like growth
factors and their receptors were most abundant in the sensory
cortex, hippocampus, and cerebellum [92].
Trophic factors such as IGF-1 promote neuronal survival and
differentiation, whereas other trophic factors (such as BDNF,
NT4/5, NT-3, via TrkB receptors) chemotactically stimulated
the migration of embryonic neurons to their cortical targets
[93].
The neonatal period is characterized by peak
rates of brain growth [94, 95] and
cortical neurons are endowed with an exhuberant production of
synaptic vesicles [96]. During the
processes of migration and differentiation, large numbers of
neurons undergo apoptosis or programmed cell death from
diverse areas of the developing cortex. Using
semiquantitative methods, Rabinowitz et al. [97] calculated that the number of neurons in
the human cortex reaches a maximumat 28 weeks of gestation,
then then declines by about 70% to achieve a stable number of
neurons at birth. The lack of afferent inputs produced by the
blockade of N-methyl-D-aspartate (NMDA) receptors in
developing neurons markedly accentuated the occurrence of
apoptosis in the neonatal rat brain with the maximun effects
occurring in 7-day-old (P7) rates [98].
Thus NMDA activity in the neonatal rat brain may be required
for cell survival and the factors that regulate apoptosis in
the neonatal brain would play important roles in the final
development of the somatosensory cortex.
The density of NMDA receptors increases
postnatally reaching peak levels of expression in the
neonatal brain [99, 100]. Whole-cell
patch-clamp analysis showed NMDA-induced Ca2+ currents
developing at P7 whereas AMPA/KA-induced currents were larger
and developed earlier [101]. Depending on
the cellular context, Ca2+ influx through voltage-sensitive
Ca2+ channels increased cell survival, whereas Ca2+ influx
via the MNDA receptors in postnatal neurons mediated
excitotoxic cell death [102-104]. The
peak susceptibility to MNDA-mediated excitotoxicity occurred
between P7 and P15, from accentuation of its metabolic
effects [105, 106] or reduction of the
voltage-dependent Mg2+ block [107],
potentially correlated with the expression of NMDA subunits
in immature neurons [108]. Permanent
alterations in the structure of NMDA receptors occurred
following exposure to neonatal hypoxic stress, resulting in
increased susceptibility to NMDA-induced toxicity thereafter
[109, 110]. NMDA receptors play a central
role in thte activity-dependent changes in dendritic length
or spine density, synaptic stability, long-term potentiation
(LTP) or depression, and other processes that mediate the
heightened plasticicy of the neonatal brain. [82, 90, 111]
NMDA-dependent mechanisms are not only
implicated in the spinal transmission of pain but also in the
long-term effects of pain (such as hyperalgesia, allodynia,
windup, and central sensitization [112])
leading to the pathogenesis of chronic pain states [113, 114]. Accumulating evidence suggests
that exposure to repetitive neonatal pain may promote an
increased susceptibility to chronic pain states mediated by
NMDA-dependent neuroplasticity [19, 115]. Although intense research activity has
focused on peripheral and spinal cord mechanisms that support
these phenonema [114, 116], we believe
that the significant role of NMDA-mediated neuroplasticity in
supraspinal areas remains unexplored. The following findings
suggest such a role.
Craig and Malenka [41]
demonstrated that projections from the ventrobasal thalamus
to layer IV neurons in the somatosensory (S1) cortex are
organized by NMDA-dependent LTP. Induction of LTP occurred in
P3-P7 rats but was absent in P8-P14 rates [41]. Further experiments showed that
NMDA-mediated LTP is responsible for the learning and memory
associated with classical fear conditioning [117, 118], and that early experiences can
modify the capacity for learning in the adult [119]. It is particularly fascinating that
the capacity for adult learning can be enhanced [119] or inhibited [120,
121] by neonatal experiences. Thus it is plausible that
the underlying templates for subsequent cognitive ability and
vulverability to psychiatric disorders may be formed as a
result of abnormal neonatal stimulation [9, 14,
15].
Critical Periods of
Vulnerability
Critical periods of
vulnerability occur just before and after birth, when the
development of the underlying neoronal circuitry is more
susceptible to pertubation than at any other time of life.
Disruptive experiences at this time may have a greater impact
on subsequent neurobiological and behavioral development [42, 47]. There is increasing clinical
appreciation for the long-term, often permanent effects of
early sensory experiences on the patterns of subsequent
behavior [20].
A search for critical periods may be aided
by the ontogenetic changes in pain threshold during early
development. Falcon et al. [77] found that
the nadir for thermal pain thresholds in infant rats occurred
at P6 and similar age-related patterns were noted in
thresholds for the cutaneous withdrawal reflex [73] and foot-shock responses [122]. When rat pups of different age groups
(P4, P7, P14) were tested in different graded stimuli in our
laboratory, the number of ultrasonic cries at P7 were
significantly increased (ANOVA, F = 57.94, p < 0.0001).
Pain behaviors occurred in response to saline injections
(0.01%) at all ages tested (P4, P7, P14). Thus, sensitivity
to painful stimulation may be maximal in rat pups at 6-9 days
of age, a period that corresponds with the neurological
maturity of full term infants at birth.
We postulate that a critical window of
neonatal brain development occurs around the time of birth in
human neonates and at 6-9 days in rat pups. This period also
coincides with the peak rates of brain growth [95] and an exuberant production of synaptic
sites [96]. At this time, the dorsal horn
of the spinal cord and other supraspinal regions show a peak
density of NMDA receptors [99, 100],
decreased thresholds for NMDA- or kainate-induced seizures
[123], an increased magnitude of Ca2+
currents from NMDA activity [107], and a
greater susceptibility of brain cells to NMDO induced
toxicity [102, 103, 124]. Neonatal brain
cells have altered molecular mechanisms for Ca2+ signalling
[102] which appear to determine neuronal
survival, whereas blockade of NMDA receptors in this period
causes widespread apoptosis [98].
Data related to normal brain development in
the neonate reviewed above support the assertion that adverse
sensory experience during the neonatal period may have
widespread and far-ranging effects on subsequent
developmental events. In the sections below, we examine the
effects of overstimulation or understimulation in neonatal
rat pups leading to permanent changes in adult behavior and
cognitive ability.
Abnormal stimulation of the Newborn
Numerous animal experiments over the past 30
years have demonstrated the long-term effects of early
experience on subsequent behavior [21,
125]. These data suggest that both excessive sensory
stimulation and the lack of appropriate stimulation in
neonatal rats may lead to permanent changes in adult rat
behavior, altered hormonal or immune responses to subsequent
stress, associated with altered expression of
neurotransmitters, their receptors, and/or other cellular
processes associated with repetitive neonatal pain and
maternal separation, used repetively as representative models
of over-or understimulation.
Longterm effects of pain in neonatal
rats
The long-term effects
of acute or repetive pain in neonatal rat pups have remained
largely unexplored. Reynolds and Fitzgerald [126] reported that skin wounds placed in
newborn rates (at P0, P7) caused a marked hyper-innervation
and decreased thresholds in the injured area lasting longer
than 3 months after the injury. Nerve sprouting from A and C
fibers was markedly increased around skin wounds place on the
day of birth (P0) as compared with older ages, and was
stimulated by nerve growth factor in older rats and unknown
neurotrophic factors in the neonate [127]. Neonatal rats subjected to
carrageenan-induced inflammation on P1 developed a persistent
expansion (by approximately 30%) in the receptive field of
the dorsal horn neurons [128], although
the mechanisms mediating these changes remain unknown.
Behavioral changes in rats following
exposure to repetitive pain as neonates were first reported
from our laboratory [129]. Rat pups
subjected to four needle pricks each day from P0 to P7 showed
greater preference for alcohol as adults, increased defensive
withdrawal and hyper-vigilance behavior. Interestingly, these
animals had diminished neuronal activity in the somatosensory
cortex following hot-plate exposure as compared to the adult
rats receiving tactile stimulation as neonates. These results
were correlated with greater exploratory and escape behavior
seen in the tactile group, whereas the experimental animal's
behavior was more reflective of fear conditioning [129]. Further experiments showed that
exposure to prolonged inflammatory pain (repetitive formalin
injections in the neonatal period) was associated with
decreased body weight, decreased pain sensitivity (hot plate
tests), and decreased locomotor activity in the adult rat.
Amongst the rats exposed to neonatal pain, adult females had
increased alcohol preference and consumption as compared to
males, consistent with our previous findings from repetitive
acute pain [129]. These preliminary
results suggest that neonatal pain leads to an increased
expression (and/or coupling) of opiod receptors in adult rats
[128], which is increased preference for
sucrose [130, 131], decreased pain
sensitivy, and the normalization of locomotor activity by
concomitant morphine treatment in neonatal rats [132]. When morphine preceded the neonatal
formalin injections, it was noted to reverse or ameliorate
the long-term effects from exposure to neonatal inflammatory
pain. Thus, the persistent behavioral changes from repetitive
neonatal pain depend on the type of stimulation (needle prick
vs. inflammation), age at the time of stimulation, concurrent
treatment with opioids, and perhaps other unexplored factors.
The neurobiological mechanisms underlying these changes are
the focus of intense reseach activity.
Longterm Effects of Maternal
Separation
Prolonged maternal
separation leads to the development of exaggerated hormonal
responses and altered neurotransmitter release, which is
qualitatively different from the effects of postnatal
handling [133, 134]. Persistent effects
of maternal separation occurred only if the daily separation
paradigm exceeded 180 min [56], leading to
elevated baseline and stress-induced plasma corticosterone
levels, and reduced efficiency of glucocorticoid negative
feedback in adult rats [56, 135]. Repetive
maternal separation in neonatal rats was associated with
accentuated stress responses, increased vulnerabity to stress
and anxiety disorders, neophobia, early development of
diabetes, hypertension, cognitive defects, and shorter life
spans in adult rats [136, 137]. Adult
rats isolated as neonates for 1 h/day from P2 to P9 had
reduced susceptibility to an amphetamine challenge [138]. Enhanced amphetamine effects were
associated with elevated dopamine levels in the ventral
stratum, possibly linked to pertubations of the dopamine
receptor systems as a result of isolation [139]. Early isolation may contribute to the
sensitization of reward mechanisms in adult animals, from
changes in the dopaminergic mesolimbocortical pathways [140, 141].
Any links between the animal experimental
data reviewed above and recent epidemiological data on the
long-term effects of perinatal or neonatal complications are
speculative at best. How do adverse neonatal experiences
lead to long-term changes in behavior and brain function?
For these relationships to be investigated rigorously, the
potential mechanisms by which the experimental findings can
be applied to explain such preliminary data must be
understood. Answers to such questions will not only help to
understand the mechanisms by which long-term effects are
produced, but will also define the clinical importance of
interventions that treat or reduce the impact of abnormal
stimulation.
Proposed Cellular Mechanisms
We propose two novel hypotheses that can be
tested in the laboratory to explain the link between
perinatal sensory experiences and adult behavior: (a)
excessive neonatal stimulation resulting from perinatal
trauma or other insults causes NMDA-mediated excitotoxicity
in multiple areas of the developing brain, and (b) lack of
appropriate sensory stimulation in the neonatal period serves
to enhance the normal occurrence of developmental apoptosis
in the neonatal brain.
Excitotoxicity Hypothesis
Prolonged pain
causes excessive activation of central afferent pain
pathways. This activation leads to release of glutamate and
other excitatory neurotransmitters from spinal and
supraspinal neurons. Binding of glutamate to metabatrophic
and NMDA receptors causes removal of the MG2+ block and
allows intracellular Ca2+ influx, ultimately leading to the
phosphorylation of second messengers and altered gene
regulation. NMDA-induced neuroplasticity is known to mediate
the wind-up and central sensitization of pain pathways [112-115]. It is evident that these phenomena
are accentuated and prolonged in the developing nervous
system [19, 101-107,
112-115], although the supraspinal
excitotoxic effects of pain-related NMDA activity have not
been investigated in immature animals.
