Diagnosis of birth asphyxia on the basis of fetal pH, Apgar

score, and newborn cerebral dysfunction

Larry C. Gilstrap III, MD, Kenneth J. Leveno, MD, Jody Burris, RN,
M. Lynne Williams, RN, and Bertis B. Little, PhD
Dallas, Texas

Imprecise diagnosis of birth asphyxia coupled with uncertainties about causal factors for neurologic
abnormalities in the newborn have greatly fueled the current litigation crisis in obstetrics. Our goal was to
more precisely define birth asphyxia based on fetal condition as measured by umbilical artery blood pH,
Apgar scores, and neurologic condition of newborns. We selected for study 2738 patients with singleton
pregnancies with cephalic presentations who were delivered of infants at term to avoid complications such
as prematurity, which may affect infant outcome independent of birth condition. The basis for study of
these particular patients were defined criteria for high risk and an indicated arterial cord pH value. A total
of five infants demonstrated cerebral dysfunction as evidenced by seizures during the neonatal period.
Infection was linked to seizures in three of these infants; one infant had neonatal asphyxia and only one
infant's clinical course could be attributed solely to birth events (uterine rupture). Stratification of umbilical
artery blood pH values, Apgar scores, and combinations of these dependent variables in relation to
newborn clinical outcomes revealed that infants must be severely depressed at delivery before birth
asphyxia can be reliably diagnosed. Such depression includes Apgar scores :s3 at 1 and 5 minutes plus
umbilical artery pH values <7.00. (AM J OBSTET GVNECOL 1989;161 :825-30.)

Key words: Birth asphyxia, acidemia, Apgar score

Perinatal and asphyxia are overused terms in current
obstetric-legal parlance. Typically these terms are used
to implicate "birth asphyxia" as the sole cause of neu-
rologically damaged infants. Imprecise definition of
these terms coupled with uncertainties about causal fac-
tors for neurologic abnormalities in the newborn have
greatly aided the current litigation crisis in obstetrics.
The term perinatal may be used to denote many dif-
ferent time periods during gestation. For example, ac-
cording to Webster's Dictionary, I perinatal refers to "the
period around the time of birth." However, this term
is defined in Dorland's Medical Dzctionary2 as "beginning
with completion of the twentieth to twenty-eighth week
of gestation and ending 7 to 28 days after birth." The
term asphyxia is also variously defined in these diction-
aries. In Webster's Dictionary, I asphyxia is defined as "a
lack of oxygen or excess of carbon dioxide that is usu.
caused by interruption of breathing and that causes
unconsciousness." According to Dorland's Medical Dic-
tionary," asphyxia is a "lack of oxygen in respired air,
resulting in impending or actual cessation of apparent
From the Divljion of Maternal-Fetal Medzcme, Department of Ob-
stetrics and G,vnecology, University of Texas Southwestern Medical
Center.
Presented at the Nmth Annual Meetmg of the Sooety of Perinatal
Obstetriciam, New Orleans, Louzsiana, February 2-4, 1989.
Reprint requests: Larry C. Gilstrap, III, MD, Department of Obstetrtc;
and Gynecology, Unzvemty of Texas Southwestern Medzcal Center.
5323 Harry Hines Blvd., Dallas, TX 75235-9032.
616113711 '
life." Other commonly used terms include hypoxia, de-
fined as low content of oxygen, or ischemia to signify
reduced perfusion.' Such variable and therefore im-
precise definitions, when applied by laymen to neu-
rologically damaged newborns, are usually accepted to
signify that the infant's outcome was a result of oxygen
deprivation during the birth process or "birth as-
phyxia."
The most common cause of hypoxia in the fetus is
hypoperfusion, or ischemia.' Many investigators now
emphasize that very recent hypoxia-ischemia can only
be implicated as a cause of brain damage when central
nervous system signs and symptoms are present in the
immediate newborn period. 3-5 Cerebral dysfunction or
encephalopathy in the newborn may be defined as a
syndrome in which tone and primitive reflexes are al-
tered from the norm and there is evidence of cerebral
irritation manifested by seizure activity." 6 Long-term
neurologic outcome is related to the severity and du-
ration of these signs in the newborn. For example, in
a study by Robertson and Finer' seizures in the im-
mediate newborn period were a marker for abnormal
neurologic sequelae.
Ideas about the obstetric antecedents of cerebral
palsy are dangerously simplistic" 7 and we sought to
more precisely define "birth asphyxia" on the basis of
both fetal condition at birth and clinical performance
of the newborn in the nursery. Our goal was to correlate
immediate neurologic outcomes of term infants with

825
 826 Gilstrap et al.
Table I. Maternal demographic in 2738 term
pregnancies with cephalic presentation and
live births
Characteristic No. % Umbtlical
Age (yr)
:519 759 28
29-29 1597 58
30-39 368 13
2:40 14 0.5
Race
Hispanic 1092 40
Black 866 32
White 741 27
Other 39 1
Nulliparous patients 1203 44
Prenatal care 2412 88
fetal acid-base status as measured in umbilical artery
blood and newborn condition as reflected by Apgar
score. Singleton term infants with cephalic presenta-
tions were selected for study to avoid complications
such as breech presentation, which may affect newborn
condition. Similarly, preterm births were excluded in-
asmuch as complications of prematurity such as respi-
ratory disease and intracranial hemorrhage may affect
infant outcome independent of birth events.
Material and methods
Between August 1, 1987, and February 28, 1988,
blood was obtained from the umbilical artery for pH
measurement in 2738 complicated pregnancies deliv-
ered at Parkland Memorial Hospital. These study par-
ticipants were chosen from a total of 8678 patients who
were delivered of infants at our institution during this
time period. The following complications were used to
identify pregnancies for umbilical blood pH determi-
nation: cesarean section, forceps delivery, meconium
in the amnionic fluid, oxytocin stimulation oflabor, and
abnormal fetal heart rate (FHR). Umbilical artery spec-
imens were drawn from doubly clamped cord segments
into 3 ml plastic syringes flushed with 1000 U / ml of
heparin solution and the specimens were transferred
in ice to the hospital laboratory for analysis. Results of
these analyses were prospectively computerized and
linked to a similarly computerized obstetric data base.
Term singleton pregnancies, defined to include birth
weights 2:2500 gm, morphologically normal infants,
and cephalic presentations, were subsequently selected
for analysis. Pediatric house officers attended all deliv-
eries and assigned Apgar scores. Low Apgar scores
were defined in this study as :56 at 1 and 5 minutes
because this score represented the criteria commonly
used to define mild asphyxia and as :53 at 1 and 5
minutes because this score repsented the criteria used
to define severe asphyxia. 8 Newborn records were sys-
tematically reviewed by research nurses with expertise
September 1989
Am J Obstet Gynecol
Table II. Results of umbilical artery blood pH
determinations in 2738 singleton live births
delivered at term
Live births
artery
blood pH No. %
2:7.25 1833 67
7.20-7.24 520 19
7.15-7.19 217 8
7.10-7.14 90 3.2
7.00-7.09 60 2.2
<7.00 18 0.6
in neonatal special care, and clinical indices of imme-
diate newborn clinical performance were prospectively
computerized. Special attep.tion was given to neurologic
signs and symptoms and their time of onset. The di-
agnosis of seizures and hypotonia was based on assess-
ments by the attending pediatricians. Only hypotonia
that persisted 24 to 48 hours after birth was considered
significant. Follow-up the infants after discharge was
not available.
Tests for statistical significance were performed by
means of X2 contingency tables or Fisher's probability
(Center for Disease Control Statistical Analysis Package
written by J. D. Erickson). Probability values :50.05
were considered significant.
Results
The maternal demographic characteristics of 2738
pregnancies that resulted in live births and were in-
cluded in this investigation are summarized in Table I.
These demographic features were similar to those of
the general obstetric population at Parkland Hospital.
Pregnancy complications in the study group reflected
criteria for umbilical artery sampling and included pro-
longed rupture of membranes (2%), chorioamnionitis
(5%), hypertension (16%), use of oxytocin (25%), ce-
sarean section (44%), meconium in the amniotic fluid
(45%), and electronic FHR monitoring (70%).
The three most common indications for cesarean sec-
tion were repeat operation (42%), dystocia (25%), and
fetal distress (18%). The majority (85%) of participants
who were delivered of infants vaginally had sponta-
neous births and the remainder were assisted by
forceps.
The results of umbilical artery blood gas determi-
nations are summarized in Table II. The majority
(86%) of infants had pH values 2:7.20 and 18 (0.6%)
had PH values <7.00.
The relation between low I-minute Apgar scores (:56
and :53) and umbilical artery blood pH is depicted in
Fig. 1. Only 143 (8%) of the 1833 newborns with an
umbilical artery pH 2:7.25 had I-minute Apgar scores
Volume 161 Diagnosis of birth asphyxia 827
Number 3

60

50

40
C

•
3 30 One mlnut. Apg• .r aco,.
Q.

20 Ii"7sl
~
10

....,..: . ....: . ...:

0 0
0
..: po; po;
N V

~
.., 0
0
po;
po; ..: 0

..:

Fig. 1. Relation between low I-minute Apgar scores and umbilical artery blood pH.

30

20

i
l
10

.......:,.. ....: ....: -.. ..

0
0
..: 8
..: ..:

....:
v
0 ... 0
0
..: 0 ..:
..:
Fig. 2. Relation between low 5-minute Apgar scores and umbilical artery pH.

~6. In contrast, 50% (n = 9) of those with a pH <7.00
had such Apgar scores at 1 minute (p < 0.05). There
was a similar relation for I-minute Apgar scores ~3
(24 of 1833 or 1.3% versus 5 of 18 or 28%; p < 0.05).
Shown in Fig. 2 is the distribution of these Apgar scores
at 5 minutes according to pH. Apgar scores ~6 at 5
minutes occurred in 10 of 1833 (0.5%) infants with pH
~7.25 compared with 5 of 18 (28%) with a pH <7.00
(p < 0.05). The results for Apgar scores <3 at 5 min-
utes were 0 of 1833 with pH ~7.25 versus 3 of 18 (17%)
with pH <7.00 (p < 0.05). Apgar scores :53 at 10 min-
utes after birth were assigned to two infants and both
had an umbilical artery blood pH <7.00.
Neonatal complications in relation to umbilical artery
blood pH are summarized in Table III. With the ex-
ception of meconium aspiration syndrome, neonatal
complications increased as the umbilical artery blood
pH decreased and those with a pH <7.00 (n = 18) had
singificantly more morbidity than did those in any other
group. One third of these 18 infants required intuba-
tion, and the incidence of respiratory disease and hy-
potonia was 28% and 17%, respectively.
A total of five infants had seizures and these are
summarized in Table IV, Four of these infants had
seizures within the first 24 hours after birth and one
convulsed on day 10. Three of the four neonates with
early seizures had an umbilical artery blood pH <7.15
and two of these had a pH <7.00. Both of these infants
 828 Gilstrap at al. September 1989
Am J Obstet GynecoI

Table III. Selected neonatal complications in relation to umbilical artery blood pH values
Umbzlical artery blood pH value

>7.25 7.20-7.24 7.15-7.19 7.10-7.14 7.00-7.09 <7.00

(n = 1833) (n = 520) (n = 217) (n = 90) (n = 60) (n = 18)

Neonatal complications No. I % No. I % No. I % No. I % No. I % No. I %

Intubation in delivery room 13 0.7 6 1 2 1 4 4t 2 3 6 33:j:
Respiratory disease* 45 3 16 3 13 6§ 6 711 5 8~ 5 28#
Meconium aspiration syn- 4 0.2 1 0.2 1 0.5 2 2 1 2 0
drome
Hypotonia 35 2 10 2 2 0.9 4 4 4 7** 3 17tt

*Ambienit oxygen administered by either halo or mechanical ventilation.

X2 Statistical comparison to pH group 2!7.25:
tOdds ratio 6.5, p < 0.007
:j:Odds ratio 70.0, p < 0.0001
§Odds ratio 2.5, p < 0.006
IIOdds ratio 2.8, p < 0.03
~Odds raio 3.6, p < 0.02
#Odds ratio 15.3, p < 0.0001
**Odds ratio 3.7, p < 0.03
ttOdds ratio 10.3, p < 0.005

had metabolic acidemia, defined as a bicarbonate level
~17.3 mEq/1 and a Peo2 :565 mm Hg. 9 All four of the
infants with early seizures had I-minute Apgar scores
:53 and 5-minute scores :56. As shown in Table IV, two
of the infants with early seizure activity had neonatal
complications (group B streptococcal sepsis and syph-
ilis) associated with such neurologic dysfunction.
Inasmuch as Apgar scores and acid-base status may
be dependent variables in the determination of the
newborn condition, we analyzed combinations of these
indices and the results are summarized in Table V.
None of the 13 neonates with a pH <7.00 and a 1-
minute Apgar score >3 had either seizures or hypo-
tonia. There were 12 of these 13 neonates who had
either respiratory or a mixed pattern of acidemia. 9 Con-
versely, among those with pH values ~7.00 and an
Apgar score :53, seven (14%) had hypotonia and
two (4%) had a seizure. One infant had an umbil-
ical artery blood pH of 7.11 and early onset group
B streptococcal sepsis. Importantly, when infants had
both low Apgar scores and pH values the frequency
of serious neonatal morbidity significantly increased
(P < 0.05). Specifically, all required resuscitation at
birth and approximately one half demonstrated neu-
rologic dysfunction.
Comment
Both the American College of Obstetricians and
Gynecologists lO and the American Academy of Pediat-
rics l l have challenged use of the Apgar score alone to
define birth asphyxia. As both of these organizations
have emphasized, Apgar scores are unsatisfactory for
the prediction of neurologic outcomes of newborns. It
is important to recognize that there are several other
factors that can account for low Apgar scores besides
"asphyxia" such as gestational age, maternal medica-
tions, type of anesthetic administered, and the person
who assigned the score. 9 In an attempt to more objec-
tively define immediate newborn condition, several in-
vestigators have recommended use of umbilical blood
acid-base determinations. 9.!2.!' However, reliance solely
on newborn acid-base status to define birth asphyxia
has been thwarted by the unrealistically high pH values
used to define pathologic acidosis. For example, a com-
monly used definition of newborn acidosis or "acide-
mia" is umbilical blood pH <7.20, but the majority of
such newborns wil be vigorous at birth and will not
have obvious neonatal sequelae. 9 .!5 This was borne out
in our investigation; 385 infants (14%) were acidotic by
this definition and yet the majority were vigorous at
birth and without neurologic sequelae.
It also has been reported that there is a poor cor-
relation between Apgar scores and umbilical blood pH
values. Sykes and associates!2 defined pathologic aci-
dosis as pH <7.11 and reported that 81 % of infants
with low 5-minute Apgar scores «7) were not acidotic
at birth; conversely, 73% of those who were acidotic
had normal Apgar scores. Regardless, the Apgar scor-
ing system continues to be the most commonly used
method to define birth asphyxia. for example, accord-
ing to The International Classification of Disease," mild as-
phyxia is defined as a I-minute Apgar score :56 and
severe asphyxia is defined as a I-minute Apgar score
:53. If these definitions were used in the present study,
Volume 161 Diagnosis of birth asphyxia 829
Number 3

Table IV. Characteristics and neonatal outcome of five infants with seizures
Umbtlical
Day of artery Apgar
Infant seizure blood pH scores* ComplicatIOns Outcome

1 10 7.31 9.9 rH isoimmunization, meningitis Discharged day 18

2 1 7.11 3, 5, 7 Meconium aspiration, group B streptococcal Died day 2
sepsis
3 6.59 2, 3, 4 Congenital syphilis Discharged day 11
4 7.20 3, 6, 8 Meconium aspiration, pneumothorax, pneumo- Died day 22
mediastinum
5 6.59 1,3,3 Uterine rupture Discharged day 30;
cerebral palsy

*Scores at 1 minute, 5 minutes, and 10 minutes.

Table V. Neonatal morbidity in relation to combination of low I-minute Apgar scores and low umbilical
artery blood pH values
pH <7.00, pH 2:.7.00; pH <7.00
Apgar score Apgar score Apgar score
>3 $3 :53
(n = 13) (n = 49) (n = 5)
Newborn compltcatlOns No.
-'
% Companson* No.
I % Comparison* No.
1 %
Intubation 1 8 NS 8 16 0.0004 5 100
Admission to special care nursery 1 8 NS 15 31 NS 2 40
Respiratory disease 2 15 NS 13 27 NS 3 60
Hypotonia 0 NS 7 14 0.03 3 60
Seizures 0 NS 2 4 0.03 2 40

*Statistical comparison of adjacent groups shown in table.

267 or approximately 10% of the newborns would have
been diagnosed to have mild asphyxia and 55 (2%) to
have severe asphyxia. Yet virtually all of these infants
were normal in the immediate newborn period.
Because neither the Apgar score nor acid-base status
alone correlated well with neonatal morbidity, we an-
alyzed the effects of combined low Apgar scores and
low pH values. Of the newborn who had an Apgar score
:53 and an umbilical artery blood pH <7.00, all resus-
citation at birth and approximately one half had neu-
rologic abnormalities that included hypotonia or sei-
zures. No infant with an Apgar score >3 and a pH
<7.00 had seizures. One infant with a I-minute
Apgar score of 3 and a pH of 7.11 had seizure ac-
tivity in the first 24 hours of life. This infant
had early onset group B streptococcal sepsis, which
likely was the cause of hypoperfusion leading to
brain damage and seizures. The combination of
an Apgar score :53 and pH <7.00 was a sensi-
tive predictor of all cases of serious neonatal mor-
bidity.
Cerebral dysfunction as evidence by seizures in the
newborn period may result from metabolic abnormal-
ities (e.g., hypoglycemia), infection, malformations,
trauma, or true birth asphyxia. 3 There were five (1.8
per 1000) infants in the present study who had signif-
icant neurologic sequelae (i.e., seizures) during the neo-
natal period. These five infants demonstrate the plu-
rality of causes for newborn cerebral dysfunction. One
newborn had a seizure on day 10 that was due to men-
ingitis. Infection (group B streptococcal sepsis and con-
genital syphilis) was implicated in two other infants.
The fourth infant had an umbilical artery blood pH of
7.20 at delivery and pulmonary insufficiency developed
as a result of meconium aspiration. The pH of this
infant decreased acutely to 6.91 soon after arrival in
the nursery and the infant subsequently convulsed.
This time sequence suggests neonatal asphyxia as op-
posed to birth asphyxia. The fifth infant had an um-
bilical artery blood pH of 6.59, 1- and 5-minute Apgar
scores of 1 and 3, respectively, and was delivered by
cesarean section because of fetal distress due to uteripe
rupture. This is the only infant in the entire study pop-
ulation who had seizures and no other associated caus-
ative factors for this brain disorder. Cerebral palsy was
diagnosed in this infant before discharge.
It is evident from this investigation of 2738 term
pregnancies with intrapartum complications that new-
830 Gilstrap et al.
born cerebral dysfunction manifested as seizures and
attributable to true birth asphyxia is extremely rare.
The study results now reported suggest the infant must
be severely depressed at birth to implicate birth as-
phyxia as the sole cause of newborn seizures. Such
depression includes (1) extremely low Apgar score :53,
(2) umbilical artery blood pH <7.00, and (3) need for
resuscitation at birth. Towbin 4 has also concluded that
infants who sustain central nervous system damage are
severely depressed at birth. He stated, "It is implausible
to believe that cerebral palsy developing in an infant
can be attributed to massive acute cerebral damage in-
curred intranatally in an infant who was in good clinical
condition soon after delivery." Other also have ob-
served that seizures within the first day of life may be
a sign of newborn brain dysfunction as a result of as-
phyxia.' 16-18 This observation that newborn brain dis-
orders can only be attributed to birth asphyxia when
the infant is extremely ill at birth is also supported by
animal studies. Such studies indicate that animals must
have almost lethal hypoxia before encephalopathy and
neurologic damage develop.3. 19We conclude that birth
asphyxia that leads to newborn seizures is very rare and
that such neurologic dysfunction may result from a
plurality of causes. Caution must be exercised before
attributing birth asphyxia as the sole explanation for
seizures in the immediate newborn period. An infant
who is asphyxiated at birth must be severely depressed
to include Apgar scores :53, fetal pH <7.00, require
resuscitation, and have seizures in the first day after
birth. The impact of lesser degrees of asphyxia as mea-
sured by Apgar score and acid-base status at birth re-
mains unknown. A limitation in the design of this study
is that there is now follow-up beyond the newborn pe-
riod and it is possible that other infants in the study
group would later show signs of cerebral dysfunction.
Thus long term follow-up studies are necessary to com-
pletely elucidate the role of birth events in the genesis
of infant brain disorders. Until such information be-
comes available, we believe it is inappropriate to di-
agnose birth asphyxia unless the infant is both severely
depressed and acidotic at birth.
September 1989
Am J Obstet Gynecol
REFERENCES
1. Webster's ninth new collegiate dictionary. Springfield,
Massachusetts: Merriam-Webster, Inc., 1985.
2. Dorland's illustrated medical dictionary, 26th ed. Phila-
delphia: WB Saunders, 1985;129.990.
3. Freeman JM, Nelson KB. Intrapartum asphyxia and ce-
rebral palsy. Pediatrics 1988;82:240-9.
4. Towbin A. Obstetric malpractice litigation: the patholo-
gists view. AMJ OBSTET GYNECOL 1986;155:927-35.
5. Robertson C, Finer N. Term infants with hypoxic-
ischemic encephalopathy: outcome at 3.5 years. Dev Med
Child Neurol 1985;27:473-84.
6. O'Brien MJ, Ash JM, Gilday DL. Radionuclide brain-
scanning in perinatal hypoxial ischemia. Dev Med Child
NeuroI1979;21:161-5.
7. Hey E. Fetal hypoxia and subsequent handicap: the prob-
lem of establishing a causal link. In: Chamberlain GVP,
Orr CJB, Sharp F, eds. Litigation and obstetrics and gy-
necology. London: Royal College of Obstetricians and Gy-
naecologists, 1985:223-42.
8. The international classification of diseases, 9th revision,
clinical modifications, 2nd ed. Washington, D.C.: 1980;
DHHS publication no. (PHS)80-1260.
9. Gilstrap LC, Hauth JC, Hankins GDV, Beck AW. Second
stage fetal heart rate abnormalities and type of neonatal
acidemia. Obstet Gynecol 1987;70:193-5.
10. American College of Obstetricians and Gynecologists
Committee on Obstetrics (maternal-fetal medicine). Use
and misuse of the Apgar score (ACOG committee on the
fetus and the newborn). Washington, D.C.: Nov 1986.
11. American Academy of Pediatrics. Use and abuse of the
Apgar score. Pediatrics 1986;7:1148-9.
12. Sykes GS, Johnson E, Ashworth F. Do Apgar scores in-
dicate asphyxia? Lancet 1982;1:494-6.
13. Silverman F, Suidan J, Wasserman J, et al. The Apgar
score: is it enough? Obstet Gynecol 1985;66:331-6.
14. Dijxhoorn MJ, Visser GHA, Fidler VJ, et al. Apgar score,
meconium, and acidaemia at birth in relation to neonatal
neurological morbidity in term infants. Br J Obstet Gy-
naecol 1986;93:217-22.
15. Suidan JS, Young BK. Acidosis in the vigorous newborn.
Obstet Gynecol 1985;65:361-4.
16. Nelson KB, ElienbergJH. Neonatal signs as predictors of
cerebral palsy. Pediatrics 1979;64:225-32.
17. Mellits ED, Holden KR, Freeman JM. Neonatal seizures.
II. A multivariate analysis of factors associated with out-
come. Pediatrics 1982;70:177-84.
18. Nelson KB, Ellenberg JH. Antecedents of cerebral
palsy: multivariate analysis of risk. N Engl J Med 1986;
315:81-6.
19. Meyers RE. Experimental models of perinatal brain dam-
age: relevance to human pathology. In: Gluck L, ed. In-
trauterine asphyxia and the developing fetal brain. Chi-
cago: Yearbook Medical Publishers, Inc., 1977:37-97.