Neonatal Encephalopathy
Following Fetal Distress
A Clinical and Electroencephalographic Study
Harvey B. Sarnat, MD, Margaret S. Sarnat,
\s=b\ Twenty-one neonates of over 36
weeks' gestation suffered perinatal as-
phyxia but not chronic hypoxia. Three
clinical stages of postanoxic encephalop-
athy were distinguished. Stage 1 lasted
less than 24 hours and was characterized
by hyperalertness, uninhibited Moro and
stretch reflexes, sympathetic effects, and
a normal electroencephalogram. Stage 2
was marked by obtundation, hypotonia,
strong distal flexion, and multifocal sei-
zures. The EEG showed a periodic pattern
sometimes preceded by continuous delta
activity. Infants in stage 3 were stuporous,
flaccid, and brain stem and autonomic
functions were suppressed. The EEG was
isopotential or had infrequent periodic
discharges.
Infants who did not enter stage 3 and
who had signs of stage 2 for less than five
days appeared normal in later infancy.
Persistence of stage 2 for more than
seven days or failure of the EEG to revert
to normal was associated with later
neurologic impairment or death.
(Arch Neurol 33:696-705, 1976)
Accepted for publication May 3, 1976.
From the departments of neurology and pedi-
atrics (neonatology), St Louis University School
of Medicine and Cardinal Glennon Memorial
Hospital for Children. The authors are now with
Princess Margaret Hospital for Children, Perth,
Western Australia.
Reprint requests to Section of Neurology, Prin-
cess Margaret Hospital for Children, Perth,
Western Australia 6001 (Dr H. Sarnat).
MD
Ischemic-anoxic encephalopathy is
the most common neurologic dis
ease of the perinatal period, and is a
major cause of chronic disability in
childhood. A variety of changing clin
ical and electroencephalographic man
ifestations occur in the hours and days
subsequent to disturbances in placen
tal or fetal circulation and lacl? of
oxygen during labor or at the time of
delivery. The effects of ischemia and
hypoxia may not be identical, but they
are difficult to separate clinically, and
both factors probably contribute to
encephalopathy in distressed fetuses.
Clinical evaluation is further impeded
by secondary postnatal complications,
such as hypoglycemia and hypocal-
cemia, which are closely related to and
not independent of perinatal asphyx
ia.
This
study attempted to provide a
systematic clinical approach to the
identification of those transitory neu
rologic signs that appear sequentially
following asphyxia near or at the time
of birth, and to suggest a relation of
the duration of these signs to progno
sis. The severity of a perinatal insult
is difficult to quantitate, but the post
natal course of the infant, together
with EEG changes, appear to offer the
best indication of later neurologic
impairment.
METHODS
Twenty-one infants were studied (Table
1). All were over 36 weeks' gestational age
at birth and none had evidence of congen
ital heart disease, traumatic cerebral inju
ries, hydrocephalus, or infection. Each
infant had a well-defined episode of fetal
distress or an Apgar score of 5 or less at
one or five minutes after delivery. In no
case could the infant's clinical state be
attributed to drugs administered to the
mother. Infants with meconium aspiration,
respiratory distress syndrome, or other
illness resulting in chronic, continuing
hypoxia were excluded. Respiratory assis
tance was required only in some of the
most severely involved infants, but re
peated determinations of blood Po2, Pco2,
and pH indicated adequate ventilation.
Infants with transitory hypoglycemia or
hypocalcemia were not excluded because
these common metabolic alterations are
considered part of the neonatal postanoxic
syndrome.
Glucose, calcium, magnesium, and elec
trolyte levels in serum were determined
within an hour after admission. Infants
unable to maintain adequate oral intake
were fed by gavage or treated with intra
venous fluids for a few days. Antibiotics
were administered to a few infants
because of suspected sepsis, but all blood
cultures were negative. Serial microbili-
rubin determinations were made in icteric
infants. Seizures were treated with phno
barbital administered intravenously every
12 hours for a total daily dose of 10 mg/kg
of body weight. The daily dose was
decreased by half after the infant had
remained free of seizures for 24 to 48
hours. Phnobarbital was withheld from
infants without seizures.
Serial neurologic examinations were per
formed at 12 to 24 hourly intervals for the
first six days, and every other day there
after until hospital discharge, usually at 2
to 3 weeks of age, or until death in a few
cases. Neurologic examination included
evaluation of level of consciousness, cor
neal and gag reflexes, autonomie func
tions, cranial nerves, muscle tone and
activity, tendon stretch reflexes, tactile
reflexes, suck, Moro, and tonic neck
responses, and oculovestibular reflex. Re
sponse to pinprick was tested first in the
pretibial region, a high-threshold zone, and
then on the plantar surface, a low-thresh
old zone, on each side. The circumference of
the head was measured and the fontanelles
palpated. Ocular fundi were examined.
Electroencephalograms were recorded
using the International 10-20 System of
electrode placement, with modified mon
tages that doubled the interelectrode
distances used in routine adult recordings.
A record was made as soon as possible after
admission to hospital, again in 24 hours,
and then every three to five days until the
infant was discharged. Both wakefulness
and sleep were sampled when the infant's
level of consciousness allowed this distinc
tion. One EEG channel was devoted to
electrocardiographic monitoring during
most of the recording.
Follow-up neurologic examinations and
EEGs were performed at 3, 6 or 9, and 12
months of age, or more frequently in a few
cases. The criteria for "normal" at a given
based on the achievement of
age were
skills within a month of the 50th percentile
in the Denver Developmental Screening
Test for Infants, with appropriate allow
ances made for infants born at less than 40
weeks' gestation. Additional criteria for
normal were the absence of hemiparesis or
other focal neurologic deficits, the presence
of normal muscle tone and reflexes, and a
normal EEG.
RESULTS
Clinical Features
Three clinical stages of neonatal
encephalopathy were distinguished,
and are summarized in Table 2. Of the
21 infants, only seven exhibited
features of stage 1, all had character
istics of stage 2, and six infants
passed through stage 3. The duration
of stage 1 ranged from Va to 18 hours.
Stage 2 lasted a mean of 4.7 days,
excluding two infants who showed
features of this stage for three and
nine weeks. Duration of stage 3
ranged from a few hours to four
weeks. Recovery from stage 2 was
heralded by disappearance of myoclo
nus and of parasympathetic discharge
(defined below), increased alertness,
an improved suck, and reversion of the
EEG from a periodic pattern to one of
continuously similar activity in wake
fulness. Improvement in the EEG
preceded the resolution of clinical
signs of stage 2.
An attempt was made to define
level of consciousness by as many of
Plum's and Posner's criteria1 as are
applicable at this age. "Hyperalert-
ness" described infants in full wake
fulness who indeed did not sleep for
up to 24 hours. The eyes were widely
open and the infants appeared to be
staring, with a decreased frequency of
blinking. Visual tracking responses
were not enhanced, however. Sponta
neous motor activity was normal or
decreased, but they had a lowered
threshold to all types of minimal
stimuli. A Moro reflex was easily
elicited with light tactile stimuli to the
trunk, a bright light, a sudden sound
of moderate loudness such as a voice,
or even odors placed under the
nostrils. It also occurred seemingly
spontaneously, in response to proprio
ceptive stimuli associated with the
infant's own sudden movements.
"Lethargy" was defined as a state
of slightly delayed but complete
responses to stimuli, a slightly in
creased threshold for eliciting such
responses, and a decreased amount of
spontaneous movements. "Obtunda-
tion" was the term applied to infants
with delayed and incomplete re
sponses, a markedly increased thresh
old to all sensory stimuli, and little or
no motor activity. Lethargic and
obtunded infants also were irritable
when disturbed. "Stuporous" infants
were those who responded only to
strong, noxious stimuli by withdrawal
of the extremity or by assuming a
decerebrate posture, had absent cor
neal and gag reflexes, and who often
required assisted respirations because
of shallow, ataxic breathing with
frequent apneic spells. Actual coma
was rare in the newborn. Spinal with
drawal reflexes were preserved in
even the most severely depressed
infants.
Rhythmic segmental myoclonus
was seen as a coarse 4- to 7-hertz
tremor occurring in an extremity or
occasionally involving facial muscles
asymmetrically without deviation of
the head and eyes. The tremor was
generated in proximal muscles and
usually was not synchronous on the
two sides, but continually shifted
focus. It was enhanced by sudden
movement of an extremity. A light
tap on the patellar ligament or
brachioradialis tendon often elicited a
clonic reflex that lasted several
seconds and increased in amplitude
before decreasing.
Characteristic postures were ob
served. The distal joints, fingers and
toes, maintained the most consistent
ly abnormal postures of strong flex
ion, and extension of the fingers was
incomplete when they were stroked on
the dorsal surface. The thumbs were
strongly flexed, adducted, and op
posed across the palms (ie, "cortical
thumbs"). The wrists also were often
flexed. These postures were enhanced
by any type of stimulation. They were
strongest in stage 2. Decerebrate
posturing was seen only in stage 3,
and usually was intermittent and
precipitated by nonspecific stimuli.
The spine extended in opisthotonos.
The elbows extended and the wrists
pronated. The hips abducted rather
than extended as they do in decere
brate adults, a feature probably more
related to anatomic differences be
tween the hips of newborn infants
and adults than to differences in brain
stem mechanisms.
Changes in autonomie function con
stituted some of the most dramatic
contrasts between the stages. Stage 1
was accompanied by generalized acti
vation of the sympathetic nervous
system, manifested by mydriasis,
tachycardia, and increased alertness.
Stage 2, in contrast, was characterized
by generalized parasympathetic ef
fects. Pupils were tiny even in dim
light. The heart rate was abnormally
slow, with only occasional increases to
over 120 beats per minute. Secretions
were often copious, obstructing respi
ratory pathways and requiring fre
quent suctioning. Gastrointestinal
 Patient/Age, Birth Fetal
motility increased and peristalsis
was wk*/Sex Weight, gm Delivery Distress
could be heard over the abdomen. 1/40/.F 3,175 Breech Not documented;
Some infants passed large, green, infant required
intubation at birth
watery stools. Stage 3 appeared to be 2/37/M 2,420 Spontaneous Abruptio placentae;
accompanied by suppression of both vaginal, vertex
presentation
infant required
intubation at birth
autonomie systems. The pupils were
3/40/M 2,835 Spontaneous Not documented; infant
small to midposition, often unequal, vaginal, vertex required intubation
and responded poorly to light. Infants presentation at birth

in stage 3 also lost body temperature Emergency Fetal tachycardia;

4/38/M 2,948
regulation and were often hypother cesarean section meconium-stained
amniotic fluid;
mie despite external warming. umbilical cord
Seizures occurred in about half of around infant's neck
the patients and were almost exclu 5/42/M 4,139 Spontaneous Fetal bradycardia;
vaginal, vertex meconium-stained
sively limited to stage 2. None oc presentation amniotic fluid;
head in vagina two
curred in stage 1 and only rarely were hours; infant required
they seen in stage 3. The seizures intubation at birth
occurred on the first or second day 6/39/M 3,204 Precipitous Head delivered
vaginal, vertex spontaneously before
after birth and in every case were presentation arrival at hospital
focal or multifocal, both clinically and
7/39/M 3,005 Elective Not documented; infant
electroencephalographically. Status cesarean section required intubation
epilepticus did not occur. The seizures at birth
were easily controlled with phnobar 8/39/F 2,892 Elective Not documented; cardiac
cesarean section arrest at one to
bital. The rate of recovery did not two hr after birth
appear to be affected by the adminis
tration of this drug, and the criteria of 9/40/F 3,005 Spontaneous Meconium-stained
vaginal, vertex amniotic fluid
stage 2 were recorded before phno presentation
barbital was given. 10/41/F 3,402 Spontaneous Meconium-stained
vaginal, vertex amniotic fluid
presentation
Electroencephalographic Features 11/44/F 2,920 Spontaneous Meconium-stained
In stage 1 the EEG was interpreted vaginal, vertex amniotic fluid;
presentation postmaturity; umbilical
as normal in wakefulness. The first cord around
infant's neck
alteration occurred early in stage 2,
12/40/F 3,742 Spontaneous Fetal bradycardia
usually in the first 24 hours. It vaginal, vertex
consisted of voltage depression of 5/xv presentation
to 25/xv activity in the delta and low- 13/40/F 3,800 Elective Not documented; infant
cesarean section required intubation
theta frequency range, and a paucity at birth
of the normal amount of upper theta 14/40/M 3,575 Emergency Fetal bradycardia;
and alpha range activity usually found cesarean section abruptio placentae;
umbilical cord around
in the normal full-term waking EEG infant's neck

(Fig 1). The low amplitude slowing 15/40/F 3,771 Elective Fetal bradycardia;
cesarean section meconium-stained
was often slightly more pronounced in amniotic fluid;
one temporal region than the other, postmaturity; umbilical
cord around infant's
but was generalized. neck; infant required
intubation at birth
A periodic pattern in wakefulness
16/40/F 2,892 Spontaneous Meconium-stained
or obtundation was seen in some amniotic fluid
vaginal, vertex
infants in stage 2 from birth, or presentation
17/37/M 3,640 Spontaneous Fetal bradycardia
appeared on the second day in most of vaginal, vertex
those who initially had low amplitude presentation
slow activity. Polymorphic sharp and
slow waves of 50 to 200 occurred 18/40/M 3,374 Elective Meconium-stained
in bursts lasting one to three seconds, cesarean section amniotic fluid
and alternated with a low-amplitude Meconium-stained
19/44/F 4,026 Spontaneous
delta and theta phase lasting three to vaginal, vertex amniotic fluid;
six seconds (Fig 2). "Premature presentation postmaturity
20/38/M 3,127 Spontaneous Fetal bradycardia;
ripples" of 16 to 20 Hz were superim vaginal, vertex meconium-stained
presentation amniotic fluid;
posed in both phases at times. The premature separation
high-amplitude phase occurred simul of placenta; umbilical
cord prolapsed and
taneously in both hemispheres, but contained hematoma
was not synchronous. 21/37/M 2,580 Breech Fetal tachycardia
 Table 1 .Perinatal Data and Outcome of Infants
Duration of Encephalopathy
^_ Abnormal
Apgar Score,
1 min/5 min Stage 1 Stage 2 Stage 3 Seizures Laboratory Data Outcome
1/4 3 days 23 days No Stuporous until death at
27 days of age

0/1 9 wk No Died at 9 wk of age

2/3 1 day 3 days Yes Spastic diplegia; no

3 wkt development; myoclonic
epilepsy at 6 mo; died
at 8 mo of age
8/5 1 to 8 days Yes Blood glucose Spastic diplegia; no
2 days level, development; myoclonic
38 mg/100 ml epilepsy; hypsarrhythmic
at 14 hr EEG; died at 11 mo of age

0/4 8 days Yes Ventricular Spastic diplegia; no

fibrillation development at 9 mo
at 2 days; of age
renal failure

1/1 8 dayst Yes Spastic diplegia; poor

development; EEG slow
dysrhythmic nonparoxysmal
at 12 mo

5 days 1 day No Poor development at

6 mo of age

5/7 6 days No Diabetic Normal at 6 mo of age

mother; no
hypoglycemia
orhypocalcemia
14hr 6 days Yes Normal at 6 mo of age

6/9 18hr 6 days Yes Blood glucose Mild developmental delay

level, 17 mg/100 ml at 9 mo of age
at 6hr
5/7 6 dayst Yes Blood glucose Spastic diplegea; slow
level, development; no seizures;
25 mg/100 ml EEG slow dysrhythmic
at 12 hr nonparoxysmal at 6 mo of age

6/3 5 days No Left renal Normal at 6 mo of age

vein thrombosis

1/4 4 days No Normal at 6 mo of age

3/. 4 days Yes Blood glucose Normal at 6 mo of age

level,
13 mg/100 ml
at 6hr
3/3 6 days No Blood glucose Normal at 6 mo of age
level, 26 mg/100 ml
at 6 hr;
Ca++, 3.8; Mg ++,
1.4 mEq/liter
at 22 hr

6/8 6hr 5 days Yes Normal at 9 mo of age

4/5 1/2 hr 4 days No Diabetic mother; Normal at 6 mo of age

blood glucose level,
33 mg/100 ml,
Ca++, 4.3, Mg ++, 1.4
mEq/liter at 2 hr
4/7 3hr 4 days Yes Blood glucose level, Normal at 9 mo of age
36 mg/100 ml
at 4 and 5 hr
7/6 18hr 4 days No Normal at 7 mo of age

1/4 3 days No Ca++, 4.0; Mg++, Normal at 6 mo of age

1.3 mEq/liter
at 12 hr

9/10 12hr 2 days Yes Ca++, 3.5 mEq/liter Normal at 12 mo

at 24 hr

*Estimated gestational age at birth.

t Recovery from stage 2 accompanied by EEG change from periodic pattern to
abnormal continuous low-amplitude delta and theta activity instead of to normal.
 Table 2.Distinguishing Features of the Three Clinical Stages of
tical gray and subcortical white mat
Postanoxic Encephalopathy in the Full-Term Newborn Infant ter, extensive neuronal loss, and gross
atrophy. No evidence of intracerebral
Stage 1 Stage 2 Stage 3 or intraventricular hemorrhage was
Level of consciousness Hyperalert Lethargic or Stuporous
obtunded found, although minimal subarach-
Neuromuscular control noid hemorrhage was present in the
Muscle tone Normal Mild hypotonia Flaccid right parietal region of patient 2.
Posture Mild distal Strong distal Intermittent Brain stem lesions were absent.
flexion flexion decerebration
Stretch reflexes Overact i ve Overact i ve Decreased or absent In an additional patient (No. 4),
Segmental myoclonus Present Present Absent myoclonic epilepsy and hypsarrhyth-
Complex reflexes mia developed at 4 months of age, and
Suck Weak Weak or absent Absent he died at 11 months of age. Necropsy
Moro Strong; low Weak; incomplete; Absent
was not performed, but pneumoen-
threshold high threshold
Oculovestibular Normal Overact i ve Weak or absent cephalography at 5 months of age
Tonic neck Slight Strong Absent showed extensive gyral atrophy, en
Autonomie function Generalized Generalized Both systems larged ventricles, and a widened suba-
sympathetic parasympathetic depressed rachnoid space.
Pupils Mydriasis Miosis Variable; often
unequal; poor A bulging fontanelle or abnormal
light reflex increase in the circumference of the
Heart rate Tachycardia Bradycardia Variable head did not occur in any patient,
Bronchial and either during the neonatal period or
salivary secretions Sparse Profuse Variable
Gastrointestinal later.
motility Normal or Increased; Variable A good outcome was seen in infants
decreased diarrhea
Seizures
in whom the clinical signs of stage 2
None Common; focal Uncommon
or multifocal (excluding and EEG reverted to normal within
decerebration) five days. The presence or absence of
Electroencephalogram Normal (awake) Early: low-voltage Early: periodic
findings continuous delta pattern with easily controlled seizure activity in the
and theta. isopotential phases. first two days did not seem to
Later: periodic Later: totally
pattern (awake). isopotential influence the prognosis. All infants
Seizures: focal who had normal development when
1-to1]4-Hz
spike-and-wave seen at 6 to 12 months of age also had
Duration Less than 24 hr Two to 14 days Hours to weeks a normal EEG at that time.
The following features of the neo
natal EEG appeared to be associated
The first change in stage 3 was recovering from a totally isopotential with a poor prognosis: (1) totally
further depression of the periodic EEG, low-voltage delta and theta isopotential record at any time; (2)
pattern. The high-amplitude bursts activity returned in wakefulness, but periodic pattern with an isopotential
decreased in frequency to every 6 to isopotential activity persisted in sleep phase between bursts of activity that
12 seconds, and the low-amplitude during the same recording. occurred less often than every six
intervals underwent further voltage Focal motor seizures were charac seconds, in wakefulness, obtundation,
depression to approach isopotential. A terized by 1- to lV2-Hz sharp and slow or stupor; (3) periodic pattern in wake
similar EEG pattern was seen during wave complexes of one hemisphere, fulness with continuous cortical activi
sleep in infants of stage 2 who had a usually in the central or temporal ty between bursts that occurred every
periodic pattern in wakefulness. With region (Fig 5). The focus sometimes three to six seconds, but that persisted
progressive involvement in stage 3, shifted to the other hemisphere a few for seven days or more; (4) recovery
the EEG became totally isopotential minutes later. Generalized paroxys from periodic pattern was to abnor
(Fig 3). mal activity and secondary bilateral mal low-amplitude slow activity in
During recovery, the emergence synchrony were not seen. Clonus and stead of to normal.
from stage 2 was heralded by the movement artifacts were easily dis The results of this study thus
reappearance of continuous cortical tinguished from true focal seizure suggest that the early prognosis of
activity without periodicity in wake activity (Fig 6). neonates suffering perinatal anoxic
fulness. If recovery from stage 2 encephalopathy is good if they do not
occurred within five days, the EEG Follow-up enter stage 3 and if the total duration
reverted to normal (Fig 4). Later Four severely affected infants died, of stage 2 is less than five days,
recoveries were characterized by a and necropsy was performed in two followed by full recovery both clini
change from the periodic pattern to (patients 1 and 2). In both cases, cally and electroencephalographically.
one of continuous low-amplitude delta extensive encephalomalacia of the The short-term follow-up precludes
and theta activity, similar to that seen cerebrum and cerebellum was seen, definitive conclusions about neuro
on the first day in some infants. In with microcystic degeneration of cor- logic sequelae beyond the first year.
Fig 1.Abnormal EEG during wakefulness showing generalized delta and theta slowing of low voltage, with
paucity of faster theta and alpha range activity (compare with Fig 4). International 10-20 System of electrode
placement. Calibration: 50/xv, 1 second.
Fig 2.Periodic EEG of patient 18 at 2 days of age while awake, during stage 2. High amplitude
bursts occur simultaneously but not synchronously in two hemispheres. Calibration: 50/xv, 1
second.
 Fig 3.Continuously isopotential EEG
COMMENT
None of the clinical or electroen-
cephalographic characteristics here
described are specific for postanoxic
encephalopathy, but rather are mani
festations of a state of functional
impairment of the brain, especially
the cerebral cortex, and also occur in
other perinatal encephalopathies.
Many sympathomimetic manifesta
tions of stage 1 may represent release
of epinephrine by the adrenals rather
than overactivity of sympathetic neu
rons or inhibition of the parasympa-
thetic system. Plasma catecholamine
levels rise dramatically in anoxic and
postanoxic states in the newborn
infant,23 and the increase in circula
ting epinephrine may be as much as
eightfold following placental insuffi
ciency.2 Hypoxia of mild degree (expo
sure to an atmosphere containing 10%
oxygen for five minutes) in the post
natal period in full-term neonates did
not result in increased urinary cate-
of stuporous infant at 5 days of age,
cholamine excretion, however.4 Hy-
poglycemia occurs commonly in post
anoxic states in neonates,5 and one
third of infants in the present study
had blood glucose concentrations of
less than 40 mg/100 ml in the first few
hours after birth, during clinical stage
1, or during early stage 2. Hypogly-
cemia does not decrease the epineph
rine content of the adrenal medulla in
newborn rats,6 and it is likely that the
hypoglycemia that follows perinatal
asphyxia is secondary to high plasma
epinephrine concentration, which
causes depletion of hepatic glycogen
reserves.
Most features of stage 2 suggest
"release from inhibition" of brain
stem and spinal cord reflexes and of
vagai and other parasympathetic
nerve function. This interpretation
implies a selective vulnerability of
brain stem inhibitory interneurons, of
the cerebral cortex, or of both. An
alternative explanation to parasym-
during stage
1
3. Calibration: 50 , 1 second.
pathetic overactivity is inhibition of
the sympathetic system. Stage 3 is an
extension of the process of progres
sive cephalocaudal neuronal depres
sion throughout the neuraxis.
Because stage 1 is brief, severe fetal
distress may result in signs of stage 2
or even stage 3 already at the time of
delivery. Fetal tachycardia precedes
bradycardia in the distressed fetus,79
corresponding in utero to stages 1 and
2, respectively. Fetal monitoring also
reveals that an early sign of loss of
central nervous system control of the
heart rate is the abolition of beat-to-
beat irregularity.79 The generalized
parasympathetic discharge of stage 2
probably accounts for the passing of
meconium stool into the amniotic
fluid. Pinpoint pupils at birth, another
parasympathetic sign, do not neces
sarily signify narcotic intoxication.
The respiratory pattern of asphyx
iated neonates is incompletely charac
terized, but our observations suggest
 Fig 4.Normal EEG of patient 18 at 6 days of age while awake, after recovering from stage 2 (same
patient as Fig 2). Child now appears normal at 9 months of age. Calibration: 50 , 1 second.

Fig 5.Focal left temporo-occipital paroxysmal activity during right focal motor seizure. Characteristic 1- to 11/2-hertz spike-and-wave
complexes are seen at T3 and 01. Vertex (Cz) referential montage. Calibration: 50/xv, 1 second. EKG indicates electrocardiogram.
 Fig 6.Movement and clonus artifacts are distinguished frqm true seizure activity
in neonatal EEG. Compare with Fig 5. Calibration: 50 , 1*second.
that central neurogenic hyperventila
tion1 and ataxic breathing1 are com
mon patterns. Apneic spells may
alternate with periods of tachypnea.
The breathing responses of fetal
lambs are not affected by bilateral
cervical vagotomy,10 indicating that
they are not under autonomie control.
They are completely inhibited, how
ever, by the maternal administration
of pentobarbital and of quantities of
thiopental sodium and methohexital
sodium equal to or less than the doses
used clinically.10 These drugs depress
cerebral activity in a manner similar
to hypoxia.
The rhythmic tremor known as "jit-
teriness," "tremulousness," or "trem
ors" is a form of rhythmic segmental
myoclonus and is found to a mild
degree in normal full-term newborns
when crying or excited. Clonus of the
jaw is especially common. The seg
mentai myoclonus of neonatal post-
anoxie encephalopathy is excessive
and is unrelated to level of conscious
ness, since it may be seen in both
stages 1 and 2 with hyperalertness
and obtundation, respectively. The
pathophysiology of this phenomenon
involves more than simply release
from inhibition of the segmental
gamma loop of monosynaptic reflexes
because it is not invariably associated
with hyperreflexia and may involve
such muscles as the orbicularis oculi as
well as those of the extremities. The
clinical distinction between segmental
myoclonus and multifocal seizures in
the newborn often is difficult to make
by observation alone; taking hold of
the clonic extremity and thus chang
ing the tension on the muscle stretch
receptors by slightly flexing or exten
ding the joint immediately arrests
clonus, but does not alter true seizure
activity, during which rhythmic con
vulsive movements continue to be felt
in the examiner's hand. The EEG also
is useful in making this distinction
(Fig 5 and 6). Neonatal segmental
myoclonus does not respond to phno
barbital.
The semiperiodic EEG pattern in
the postanoxic state has previously
been observed in full-term neo-
nates1112 and also in older children and
adults.1316 It resembles the discontin
uous EEG of premature infants of less
than 30 weeks' gestation and the
normal quiet sleep record ("trac
alternant") of normal 32- to 42-week
neonates.121719 The pattern is nonspe
cific, and in full-term newborns often
occurs as a postictal phenomenon.
Postictal depression may even ap
proach an isoelectric state, but seldom
persists as such for more than 20
seconds.12 Some authors regard the
postanoxic periodic EEG pattern as a
form of neonatal seizure activity.20 We
favor an alternative explanation: the
periodic pattern represents impair
ment of intrinsic cortical rhythms
(low-amplitude phase) and preserva
tion of cortical responsiveness to
direct activation from thalamic and
brainstem centers (semiperiodic dis
charges).
The similarity of the periodic EEG
pattern between neonates and adults
in the postanoxic state suggests that
the pathogenesis is not related exclu
sively to maturational factors. A
similar periodic EEG is in normal
seen
premature infants, probably because
of the immature state of synaptogen-
esis in the cortex. The third trimester
of gestation is a time of active prolif
eration of cortical synapses, and coin
cides with a shift from a periodic EEG
to one of continuous cortical activity
during wakefulness. Neurons in the
kitten cortex rarely exhibit sponta
neous discharges in the immediate
postnatal period,21 probably because
inhibitory synapses predominate at
this stage of cortical development.22 In
asphyxiated full-term human infants,
impairment of synapses or depletion
1. Plum F, Posner J: Diagnosis of Stupor and
Coma, ed 2. Philadelphia, FA Davis Publishing
Co, 1974.
2. Cheek DB, Malinek M, Fraillon JM: Plasma
adrenalin and noradrenalin in the neonatal
period, and infants with respiratory distress
syndrome and placental insufficiency. Pediatrics
31:374-381, 1963.
3. Holden KR, Young RB, Piland JH, et al:
Plasma pressors in the normal and stressed
newborn infant. Pediatrics 49:495-503, 1972.
4. Stern L, Ledue J, Lind J: Hypoxia as a
stimulus to catecholamine excretion in the
newborn infant: II. Effect of exposure to 10%
oxygen. Acta Paediatr Scand 53:13-17, 1964.
5. Harris RJ: Plasma nonesterified fatty acid
and blood glucose levels in healthy and hypox-
emic newborn infants. J Pediatr 84:578-584,
1974.
6. Girard JR, Zeghal N: Adrenal catecholamine
content in fetal and newborn rats. Biol Neonate
26:205-213, 1975.
7. Saling E: Foetal and Neonatal Hypoxia in
Relation to Clinical Obstetric Practice. Balti-
more, Williams & Wilkins Co, 1968, pp 121-124.
8. Hammacher K, Huster KA, Bokelmann J, et
al: Foetal heart frequency and perinatal condi-
tion of the foetus and newborn. Gynaecologia
166:349-360, 1968.
9. Hon EH: An Atlas of Fetal Heart Rate
of cortical neurotransmitters may
cause the EEG to resemble that seen
in the normal premature state of
sparse synapse formation. Hypother
mia also may play a role in potenti
ating the isopotential records seen in
stage 3.12
The results of this study suggest
that the quantitation of clinical and
EEG data in terms of the duration of
transitory findings during the first
week after birth defines the severity
of the insult to the brain and increases
the accuracy of early prognosis in
asphyxiated neonates. The persis
tence of abnormal EEG patterns after
the second week in full-term infants
already has been noted by Monod et
al23 to indicate a poor prognosis. In
experimental hypoxia in fetal sheep,
the time to recovery of normal EEG
rhythms was a function of the length
of the isoelectric stage, but not of the
total period of hypoxia.24
On the basis of our findings of
apparent tolerance of neonates to
clincal and EEG features of stage 2
for five days without severe, irrever-
References
Patterns. New Haven, Conn, Harty Press, 1968.
10. Condorelli S, Scarpelli M: Fetal breathing:
Induction in utero and effects of vagotomy and
barbiturates. J Pediatr 88:94-101, 1976.
11. Dreyfus-Brisac C, Larroche JC: Discontin-
uous EEGs in premature and full-term neonates:
Electro-anatomo-clinical correlations, abstracted.
Electroencephalogr Clin Neurophysiol 32:575,
1972.
12. Engel RCH: Abnormal Electroencephalo-
grams in the Neonatal Period. Springfield, Ill,
Charles C Thomas Publisher, 1975.
13. Pampiglione G: Electroencephalographic
studies after cardiopulmonary arrest. Proc R Soc
Med 55:653-657, 1962.
14. Hockaday JM, Potts F, Epstein E, et al:
Electroencephalographic changes in acute cere-
bral anoxia from cardiac or respiratory arrest.
Electroencephalogr Clin Neurophysiol 18:575-586,
1965.
15. Pampiglione G: Post-anoxic encephalopa-
thies in children, abstracted. Electroencephalogr
Clin Neurophysiol 32:574, 1972.
16. Prior PF: The EEG in Acute Cerebral
Anoxia. Amsterdam, Excerpta Medica, 1973.
17. Samson Dollfus D: L'electroencephalo-
gramme du prematur\l=e'\ jusqu' \l=a`\l'\l=a^\gede trois mois
et du nouveau-n\l=e'\a terme, thesis. Paris, 1955.
18. Dreyfus-Brisac C, Fischgold H, Samson
Dollfus D, et al: Vielle, sommeil et reactivite
sible brain damage, we recommend
that at least two EEG records be made
during the first week in asphyxiated
infants, one on the second day and
another on the sixth day after birth.
Both wakefulness and sleep should be
sampled. If still abnormal at 6 days of
age, additional EEG recordings should
be made every three or four days until
it reverts to normal or shows no
significant change by two weeks. The
time course of the composite clinical
and EEG changes provides a more
sound basis for prognosis than do
isolated symptoms and signs, the
presence of seizures, or EEG changes
alone.
Paul A. Byrne, MD, of Cardinal Glennon
Hospitals, lent support in carrying out this study.
Dr Byrne and Jackson K. Eto, MD, referred
many of the patients. Jeffrey Modin, MD,
assisted in the follow-up of two patients, and
Mary E. Case, MD, performed the postmortem
neuropathologic examination of two infants.
Gerald D. Pratt of Cardinal Glennon Hospital
provided technical assistance with many of the
EEG recordings. Dr Byrne, Simon Horenstein,
MD, Arthur E. McElfresh, MD, and Robert M.
Woolsey, MD, provided suggestions and com
ments.
sensorielle chez le prematur\l=e'\et le nouveau-n\l=e'\:
Activit\l=e'\electrique c\l=e'\r\l=e'\braledu nourrisson. Elec-
troencephalogr Clin Neurophysiol 6(suppl):418\x=req-\
440, 1956.
19. Dreyfus-Brisac C, Flescher J, Plassart E:
L'electroencephalogramme: Crit\l=e'\red'\l=a^\geconcep-
tionnel du nouveau-n\l=e'\a terme et prematur\l=e'\.Biol
Neonate 4:154, 1962.
20. Rose AL, Lombroso CT: Neonatal seizure
states: A study of clinical, pathological, and elec-
troencephalographic features of 137 full-term
babies with long-term follow-up. Pediatrics
45:404-425, 1970.
21. Huttenlocher PR: Development of cortical
neuronal activity in the neonatal kitten. Exp
Neurol 17:247-262, 1967.
22. Purpura DP: Synaptogenesis in mamma-
lian cortex: Problems and perspectives, in
Sterman MB, McGinty DJ, Adinolfi AM (eds):
Brain Development and Behavior. New York,
Academic Press Inc, 1971, pp 23-41.
23. Monod N, Pajot N, Guidasci S: The neo-
natal EEG: Statistical studies and prognostic
value in full-term and pre-term babies. Elec-
troencephalogr Clin Neurophysiol 32:529-544,
1972.
24. Mann LI, Prichard JW, Symmes D: EEG,
ECG, and acid-base observations during acute
fetal hypoxia. Am J Obstet Gynecol 106:39-51,
1970.