Original

Article
&

&

&

&

&

&

&

&

&

&

&

&

&

&

Prognostic Factors for Survival of Growth-Restricted Fetuses With

Absent End-Diastolic Velocity in the Umbilical Artery
Riza Madazli, MD

OBJECTIVE:
The aim of the study was to evaluate the prognostic factors for survival of
growth - restricted fetuses with absent end - diastolic velocity in the umbilical
artery.
STUDY DESIGN:
Forty - five intrauterine growth - restricted fetuses with either absent end diastolic ( 34 fetuses ) or reverse flow ( 11 fetuses ) in the umbilical artery
were studied. The clinical characteristics of these pregnancies were
determined. Logistic regression analysis was conducted to find the relative
value of gestational age at birth, thoracic artery and middle cerebral artery
Doppler velocimetry, and short - term variability determined by cardiotochography in the prediction of perinatal mortality.
RESULTS:
The mean gestational age at birth and birth weight were 30.8 2.4 weeks
and 972 337 g, respectively. The perinatal mortality was 40%. Gestational
age at birth was found to have the only significant contribution to the
prediction of perinatal deaths. Fetuses with a gestational age at delivery less
than 29 weeks died and more than 31 weeks survived. Thoracic artery
pulsatility index had the best screening efficiency for predicting perinatal
mortality between 29 and 31 weeks gestational age.
CONCLUSION:
Absent end - diastolic velocity in the umbilical artery is mainly a problem of
severe preterm growth - restricted fetuses and is associated with high
perinatal mortality. The major and dominant influence on survival is
gestational age at birth.
Journal of Perinatology (2002) 22, 286 290 DOI: 10.1038/sj/jp/7210715

INTRODUCTION
Intrauterine growth restriction is a significant contributor to
perinatal mortality and morbidity and remains a major obstetric and

Department of Obstetrics and Gynaecology, Cerrahpasa Medical Faculty, University of Istanbul,

Istanbul, Turkey.
Address correspondence and reprint requests to Riza Madazli, MD, 7 8 Ksm, L1 - D, D:30,
34750 Atakoy, Istanbul, Turkey.

neonatal challenge.1 Placental-bed biopsy studies indicate that a

failure of the placenta to create an adequate uteroplacental
circulation is the major underlying factor in the pathogenesis of fetal
growth restriction.2,3 The best treatment we can offer a fetus with
growth restriction due to uteroplacental insufficiency whose
circulation and homeostasis are progressively deteriorating is
extraction at the most appropriate moment. So the most critical
decision for such pregnancies is the appropriate timing of delivery,
which is often difficult to judge.
Absent or reversed end-diastolic flow velocity (AEDFV) waveforms
in the umbilical arteries of growth-restricted fetuses have been
presented by many investigators as a sign of poor fetal outcome.4,5
The prevalence of perinatal death in this group is reported to be over
40% and fetal blood obtained by funicentesis frequently demonstrates
evidence of chronic hypoxia.6,7 Therefore, many investigators have
suggested that perinatal management of these pregnancies should be
aggressive and AEDFV detected in the umbilical artery of a growthrestricted fetus is an indication for delivery.8,9 However, decision of
delivery is not always that easy. AEDFV in the umbilical artery is
mostly present before a gestational age of 34 weeks.10 In some cases,
depending on the decision of delivery, a fetal death may translate into
a neonatal loss. This finding in the preterm fetus presents a clinical
dilemma in the management (i.e., conservative vs. delivery).
Therefore, it will be of great importance to find out the prognostic
factors for survival for pregnancies with AEDFV in the umbilical
artery.
The aim of this study is to evaluate the clinical characteristics and
the prognostic factors for survival of growth-restricted fetuses with
AEDFV in the umbilical artery.
MATERIAL AND METHODS
The study group consisted of 45 intrauterine growth-restricted
fetuses with either absent end-diastolic velocity (34 fetuses) or
reverse flow (11 fetuses) in the umbilical arteries. Fetuses with
chromosomal or structural abnormalities were excluded from the
study. Pregnancies were precisely dated based on reliable menstrual
dates or early ultrasonography. Intrauterine growth restriction was
diagnosed when fetal abdominal circumference was more than 2 SD
below the mean for gestational age and was confirmed by the serial
assessment of the fetal growth parameters.11 All of the pregnancies
had oligohydramnios and 53.5% of them were preeclamptic. The
diagnosis of preeclampsia was established in accordance with the
Journal of Perinatology 2002; 22:286 290
# 2002 Nature Publishing Group All rights reserved. 0743-8346/02 $25

286

www.nature.com / jp

Absent End-Diastolic Velocity in the Umbilical Artery

Madazli

Table 1 Patient Characteristics and Fetal Outcome

N
Age
( mean SD, yr ) [ range ]
Parity
( mean SD ) [ range ]
Preeclampsia
( n / N, % )
Gestational age at birth
( mean SD, wk ) [ range ]
Birth weight
( mean SD, g ) [ range ]
Umbilical vein pH at birth
( mean SD ) [ range ]
Perinatal deaths
( n / N, % )

45

Table 2 Perinatal Mortality of Growth-Restricted and Appropriate for

Gestational Age Fetuses Taken as Control Group, According to the
Gestational Age at Birth and Birth Weight

29.66 6.55 [ 18, 44 ]

Intrauterine
growth - restricted fetuses

0.46 0.87 [ 0, 3 ]

n/N

n/N

Gestational age at birth ( wk )

28
7/7
29 31
11 / 24
32
/ 14

100
45.8

14 / 19
8 / 44
1 / 86

73.6*
18.1 ( p < 0.001)
1.2

Birth weight ( g )
800
13 / 13
800 1000
5 / 16
1000
/ 16

100
31.3

14 / 18
8 / 36
1 / 95

77.7 ( p < 0.001)

22.2 ( p < 0.001)
1.1

25 / 45, 55.5
30.86 2.46 [ 25, 36 ]
972 337 [ 500, 2180 ]
7.17 0.11 [ 6.91, 7.29 ]
18 / 45, 40

Appropriate for
gestational age fetuses

*p < 0.01.

definitions of the American College of Obstetricians and

Gynecologists (ACOG).12
After informed verbal consent was obtained, Doppler recordings
were performed using an Ultramark 9 (Advanced Technologies
Laboratory, Bothell, WA) with a 3.5-Mhz curved linear-array
transducer. The high-pass filter was set at 100 Hz. All
measurements were performed with the mothers in a semirecumbent position and when there were no fetal gross body or
chest movements. Umbilical artery velocity waveforms were recorded
by means of previously reported technique.13 The diagnosis of
AEDFV was confirmed by Doppler sampling obtained at three
different sites. AEDFV was defined when shifts in frequency at the
end of diastole were absent and reverse flow when there were
detectable reverse velocity flow during diastole. Following the
diagnosis of AEDFV in the umbilical artery, recordings were made
from the middle cerebral artery and thoracic aorta as previously
described.13 We attempted to achieve an angle close to 0 degrees
between the Doppler ultrasound beam and the direction of blood
flow in each vessel. The pulsatility index (PI) [(maximum
velocity minimum velocity)/mean velocity] was calculated and
the mean of five consecutive cardiac cycles was taken for each
group of waveform. Detected PI values were defined as normal,
high, or low for that gestational age according to referred specific
charts for each vessel.11
Fetal heart rate monitoring was recorded for 60 min at the
same time as the Doppler studies, using a Hewlett-Packard 8040
cardiotochograph (Hewlett-Packard, Boblingen, Germany). The
Sonicaid System 8000 (Sonicaid, Oxford, UK) was used for online computer analysis of the 60-min fetal heart rate tracings.
Tracings were identified as reactive, nonreactive, and decelerative
by the system and also verified by visual analysis. The shortterm variability calculated by the computer was used for
analysis.
Delivery decision for each pregnant woman was made
individually, taking into account the maternal condition (severe
Journal of Perinatology 2002; 22:286 290

preeclamptic patients were delivered for maternal reason),

gestational age of the fetus, cardiotocography, and Doppler
findings as a whole. The families were counseled regarding the
chances of survival based on survival rates in our hospital and they
were involved in decision making, particularly when a decision of
not to deliver was made. The time interval between the time of
diagnosis and intrauterine death was determined for each of these
stillbirths.
All of the liveborn deliveries were performed by cesarean section.
The umbilical cord was clamped within 5 seconds after delivery, and
blood samples were drawn anaerobically into heparinized polyethylene syringes and analyzed in an automatic blood gas analyzer
(ABL-3 radiometer, Kopenhag, Denmark) within 10 minutes after
delivery.
One-way ANOVA and Pearson chi-square tests were used to
compare the mean values and the frequency of outcomes between
the fetuses that died and survived. Logistic regression analysis was
performed to determine the contribution of the various
independent variables in the prediction of perinatal deaths.
Receiver operator characteristic curves were used to determine the

Table 3 Clinical Characteristics of the Growth -Restricted Fetuses

That Died and Survived

N
Gestational age at birth
( mean SD, wk )
Birth weight
( mean SD, g )
Umbilical vein pH at birth
( mean SD )

Death

Alive

18
29 1.6

27
32.2 2.2

0.000

727 156

1136 328

0.000

7.04 0.13

7.21 0.06

0.001

287

Madazli

Absent End-Diastolic Velocity in the Umbilical Artery

Table 4 Relationship of Various Independent Variables With Occurrence of Perinatal Deaths Analyzed by Overall Logistic Regression
Variable

Coefficient

Gestational age at birth

Umbilical artery PI
Thoracic aorta PI
Middle cerebral artery PI
Short - term variability

1.56
0.10
0.03
0.04
0.22

screening efficiency of various variables for predicting perinatal

mortality.
RESULTS
The clinical characteristics of the study group are given in Table 1.
Gestational age at delivery was equal to or less than 30, 32, and
34 weeks in 51.1%, 73.3%, and 91.1% of the cases, respectively. Birth
weight was equal to or less than 1000 and 1500 g in 66.7% and 95.6%
of the cases, respectively. There were only four fetuses born greater
than 34 weeks of gestation and two fetuses with birth weight more
than 1500 g. The mean umbilical vein pH at birth of the liveborn
fetuses was 7.170.11 with a range of 6.91 to 7.29. All of the liveborn
fetuses had pH values 2 SD below the mean for gestational age.11
There were 18 perinatal deaths due to fetal asphyxia. Of these
deaths, 10 were stillbirths and the remaining 8 were early neonatal
deaths. The time interval between detection of AEDFV in the
umbilical artery and antepartum death ranged between 3 and 18 days
with a median of 7 days. Perinatal mortality of the growth-restricted

1,00

,75

,50

Sensitivity

,25

MCAPI
UAPI
AOPI

0,00
0,00

,25

,50

,75

1,00

1 - Specificity

Figure 1. Receiver operator characteristic curves of umbilical artery,

thoracic artery, and middle cerebral artery PI values for the prediction of
perinatal mortality (24 fetuses with gestational age at birth between 29 and
31 weeks ). MCAPI = middle cerebral artery pulsatility index; UAPI = umbilical artery pulsatility index; AOPI = thoracic artery pulsatility index.
288

Odds ratio

Significance

0.20
1.10
0.96
0.95
0.79

0.02
0.93
0.98
0.97
0.59

R
0.27
0.00
0.00
0.00
0.00

and appropriate for gestational age fetuses taken as a control group,

according to the gestational age at birth and birth weight, is shown
in Table 2. The perinatal deaths of growth-restricted fetuses were all
less than 32 weeks of gestation and birth weights were less than
1000 g. The clinical characteristics of the growth-restricted fetuses
that died and survived are illustrated in Table 3.
Cardiotochography tracings were found to be reactive, nonreactive, and decelerative in 6 (13.3%), 27 (60%) and 12 (26.7%) of
the fetuses, respectively. The mean short-term variability of the
45 fetuses were 3.531.47 cycles. There were no significant
differences between the short-term variability of the fetuses that died
and survived (3.11.1 vs. 3.71.6, p=0.298). Of the fetuses with
reactive, nonreactive, and decelerative tracings 1 (16.7%), 13
(48.1%), and 4 (33.3%) died, respectively ( 2 =2.33, p=0.312).
The fetus with a reactive tracing that died in the early neonatal
period due to perinatal asphyxia was delivered at 29 weeks gestation
for maternal condition (severe preeclampsia). Of the fetuses with
reverse flow in the umbilical artery, absent end diastolic flow in both
the umbilical artery and the thoracic aorta and absent end diastolic
flow only in the umbilical artery 5 (45.5%), 10 (50%), and 3
(21.4%) died, respectively ( 2 =2.98, p=0.225).
To determine the relative value of gestational age at birth, umbilical
artery, thoracic artery, and middle cerebral artery Doppler velocimetry
and short-term variability determined by cardiotochography in the
prediction of perinatal mortality, we conducted logistic regression
analysis (Table 4). Gestational age at birth was found to have the only
significant contribution to the prediction of perinatal deaths. As all of
the fetuses with a gestational age at delivery less than 29 weeks died
and more than 31 weeks survived, we specifically evaluated the
mentioned period. There were 24 fetuses (53.3% of the cases) of
which 11 died (11/24, 45.8%) with a gestational age at birth between
29 and 31 weeks. Figure 1 shows the receiver operator characteristic
curves of umbilical artery, thoracic artery and middle cerebral artery
PI values for the prediction of perinatal mortality. Thoracic artery PI
value of 2.90 is found to have the best screening efficiency with 82%
sensitivity and 47% specificity for predicting perinatal mortality for the
cases delivered between 29 and 31 weeks gestation.
DISCUSSION
Inadequate uteroplacental perfusion leads to fetal hypoxemia, which
triggers the fetal defense mechanisms designed to limit the extent of
Journal of Perinatology 2002; 22:286 290

Absent End-Diastolic Velocity in the Umbilical Artery

asphyxia and to avoid fetal death. A healthy fetus is equipped with a

significant number of mechanisms and strategies for reducing the
impact of asphyxia. Severe and possibly irreversible damage occurs
only when the impact of anoxia exceeds the capacity of defense
mechanisms. The goal of any antenatal fetal surveillance method is
to detect the disease state of the fetus and prevent perinatal mortality.
The main objective of any clinician should be the detection of fetuses
that are at risk of developing severe acidemia before they become
acidotic.
The umbilical arteries reflect maldevelopment of the placental
tertiary stera villous tree, the principal site of respiratory gas
exchange.14,15 The magnitude of histologic changes correlates with
abnormal umbilical artery Doppler velocimetry, being severe in
cases of AEDFV.16 Absent end-diastolic velocity in the umbilical
artery is mostly present in severe growth-restricted fetuses before 32
to 34 weeks gestation, being a very rare finding after 36 weeks
gestation.10 In our study group, 73% and 91% of the cases are also
found to be less than 32 and 34 weeks gestation, respectively. If
there is a developmental problem of the villi, so severe to form
AEDFV in the umbilical artery, it is not surprising to detect it
clinically before 34 weeks gestation. Thus, AEDFV detected in the
umbilical artery is mainly a problem of severe preterm growthrestricted fetuses.
The finding of AEDFV in the umbilical artery is associated with
high perinatal mortality.4 We also observed a perinatal mortality
of 40%. Fetuses with AEDFV in the umbilical artery, if not
delivered, died in utero within 3 weeks (median 7 days).
Furthermore, all of the liveborn fetuses were acidemic, with pH
values 2 SD below the mean for gestational age. Therefore, we
believe, like many others, that AEDFV detected in the umbilical
artery of a growth-restricted fetus is an indication for delivery.
However, for some fetuses delivery decision may only translate a
fetal death into a neonatal loss hence the mode of death in
these circumstances depends on the action or inaction of the
obstetrician.
Our data suggest that gestational age at birth is the major
contributor to the perinatal deaths. All of the fetuses with a
gestational age at delivery less than 29 weeks died and more than
31 weeks survived. This close connection between gestational age at
birth and survival may be more than a straightforward gestational
agesurvival relation. Fetuses with the same pathologic finding as
AEDFV in the umbilical artery at different gestational ages will
probably have different placental histologies and functional
capacities. Fetuses with more severe placental pathology will have
an early detection of AEDFV in the umbilical artery and less
chance of survival. At least for our unit, severe preterm and
growth-restricted fetuses (less than 28 weeks gestation and
estimated fetal weight less than 800 g) with AEDFV in the
umbilical artery are found to have no chance of survival.
Therefore, gestational age and chance of survival of such fetuses
should be taken into consideration in decision making and
counseling with the parents.
Journal of Perinatology 2002; 22:286 290

Madazli

Gestational ages at birth between 29 and 31 weeks were the

critical period for survival in our group. More than half of the
fetuses (53.3%) were born between these gestational weeks and
almost half of them (45.8%) died (antenatal or neonatal). In this
specified time period, thoracic artery Doppler investigation is found
to have the best prognostic efficiency for survival. Increased
impedance in the thoracic artery is a sign of redistribution of fetal
blood. Thoracic artery PI values above 2 SD for gestational age is
found to have 82% sensitivity and 47% specificity for predicting
perinatal mortality. The absence of end-diastolic frequencies in the
aortic flow velocity waveforms is a sign of severe redistribution of
the fetal blood, associated with high incidence of perinatal
mortality and morbidity.17 Our data also confirm that growthrestricted fetuses with AEDFV detected both in the umbilical artery
and thoracic aorta are severely compromised, and time gained in
utero has no benefit for the fetus. Timing of delivery of fetuses with
AEDFV in the umbilical artery depends on gestational age, and
good clinical judgment with individualization is needed. Both
maternal and fetal riskbenefit assessments must be taken into
consideration.

References
1. Holmes RP, Soothill PW. Intrauterine growth retardation Curr Opin Obstet
Gynecol 1996;8:148 54.
2. Meekins JW, Pijnenborg R, Hanssens M, McFayden IR, van Asshe A. A
study of placental bed spiral arteries and trophoblast invasion in
normal and severe preeclamptic pregnancies Br J Obstet Gynecol 1994;
101:669 74.
3. Madazl R, Budak E, Calay Z, Aksu MF. Correlation between placental bed
biopsy findings, vascular cell adhesion molecule ( VCAM - 1 ) and fibronectin
levels in preeclampsia Br J Obstet Gynecol 2000;107:514 8.
4. Karsdorp VHM, van Vugt JMG, van Geijn HP, et al. Clinical significance of
absent or reversed end diastolic velocity waveforms in umbilical artery Lancet
1994;344:1664 6.
5. Satoh S, Nakano H. Clinical applications of the Doppler technique in
monitoring the fetus Clin Perinatol 1999;26:853 68.
6. Lin CC, Santolaya - Forgas J. Current concepts of fetal growth restriction: Part
II. Diagnosis and management Obstet Gynecol 1999;93:140 6.
7. Madazl R, Sen C, Uludag S, Ocak V. Doppler dynamics: their clinical
significance and relationship with fetal blood gases and pH measurements
J Obstet Gynecol 2001;21:440 4.
8. Alfrevc Z, Neilson JP. Doppler ultrasonography in high risk pregnancies:
systematic review with meta - analysis Am J Obstet Gynecol 1995;172:
1379 87.
9. Brar HS, Platt LD. Reverse end diastolic flow velocity on umbilical artery
velocimetry in high risk pregnancies: an ominous finding with adverse
pregnancy outcome Am J Obstet Gynecol 1988;159:559 63.
10. Kingdom JCP, Rodeck CH, Kaufmann P. Umbilical artery Doppler more
harm than good? Br J Obstet Gynecol 1997;104:393 6.
11. Chitty LS, Altman DG. Appendix: charts of fetal measurements. In: Rodeck
CH, Whittle MJ, editors. Fetal Medicine, Basic Science and Clinical Practice.
London: Churchill Livingstone, 1999;104:1095 140.
12. American College of Obstetricians and Gynecologists. Management of
289

Madazli

13.

14.

15.

290

preeclampsia. Washington, DC: American College of Obstetricians and

Gynecologists, Technical Bulletin No 91, 1986.
Deane C, Harrington K. A practical approach to obtaining optimum Doppler
signals. In: Harrington K, Campbell S, editors. A Colour Atlas of Doppler
Ultrasonography in Obstetrics. London: Arnold, 1995;168:35 46.
Krebs C, Macara LM, Leiser R, Bowman AWF, Greer IA, Kingdom JCP.
Intrauterine growth restriction with absent end - diastolic flow velocity in the
umbilical artery is associated with maldevelopment of the terminal placental
villous tree Am J Obstet Gynecol 1996;175:1534 42.
Macara LM, Kingdom JCP, Kaufmann P, et al. Structural analysis of

Absent End-Diastolic Velocity in the Umbilical Artery

16.

17.

placental terminal villi from growth - restricted pregnancies with abnormal

umbilical artery Doppler waveforms Placenta 1966;17:37 48.
Hitschold T, Weiss E, Beck T, Huntefering H, Berle P. Low target birth weight
or growth retardation? Umbilical Doppler flow velocity waveforms and
histometric analysis of fetoplacental vascular tree Am J Obstet Gynecol
1993;168:1260 4.
Hackett GA, Campbell S, Gamsu H, Cohen - Overbeek T, Pearce JMF.
Doppler studies in the growth retarded fetus and prediction of necrotising
enterocolitis, haemorrhage and neonatal morbidity Br Med J 1986;294:
13 9.

Journal of Perinatology 2002; 22:286 290