Expectant Management in

Pregnancies with Severe Pre-eclampsia

Bassam Haddad, MD,* and Baha M. Sibai, MD†

The main objective of expectant management in women with severe pre-eclampsia (PE)
remote from term is to improve neonatal outcome. Maternal conditions, however, may
worsen during expectant management. This highlights the importance of balancing the
risks between maternal and perinatal outcomes. Traditionally, women with severe PE
remote from term are delivered expeditiously, regardless of gestational age. We here have
reported several retrospective, case-control, observational, prospective, or randomized
trials in which expectant management in women with severe PE was feasible in well-
selected patients without prejudicing maternal safety, and we have described our rationale
and guidelines for this management.
Semin Perinatol 33:143-151 © 2009 Elsevier Inc. All rights reserved.

KEYWORDS preeclampsia, severe preeclampsia, management, expectant management

Introduction methods for monitoring maternal and neonatal status led

several authors to challenge the traditional idea that women

P re-eclampsia (PE) complicates 3% of pregnancies. It re-

mains a major cause of maternal and perinatal morbidity
and mortality, particularly in its severe forms. PE is usually
defined as severe in the presence of one of the following:
systolic blood pressure ⬎160 mm Hg or diastolic blood pres-
sure ⬎110 mm Hg, eclampsia, pulmonary edema, symptoms
suggesting significant end organ involvement (such as persis-
tent headache, visual disturbances, epigastric or right upper
quadrant pain), oliguria ⬍500 mL/24 hours, microangio-
pathic hemolysis, thrombocytopenia, severe intrauterine
growth retardation (IUGR), or oligohydramnios.1 Early se-
vere PE is associated with a progressive deterioration of the
maternal condition, and delivery remains the only definite
treatment. There is general agreement to terminate the preg-
nancy when women have severe PE beyond 34 weeks’ gesta-
tion. Delivery at earlier gestational age, however, is associated
with increased risk of adverse neonatal outcome.2 In addi-
tion, fetal lung maturity is not accelerated by PE,3 and neo-
natal outcome remains closely dependent on the use of cor-
ticosteroids for fetal lung maturity enhancement.4 Improving
*Department of Obstetrics and Gynecology, University Paris XII, Creteil,
France.
†Department of Obstetrics and Gynecology, University of Cincinnati, Cin-
cinnati, OH.
Address reprint requests to Bassam Haddad, MD, Department of Obstetrics
and Gynecology, CHI Creteil, 40, Avenue de Verdun, 94010, Creteil,
France. E-mail: bhaddad@chicreteil.fr
with severe PE needed to be delivered immediately. Because
of improvement of neonatal outcome after corticosteroid
prophylaxis, many investigators postponed delivery for 48
hours to allow a complete course of corticosteroid prophy-
laxis. In addition, some of these women had disease stabili-
zation during the first 48 hours, and therefore pregnancies
were prolonged.
Expectant management, however, may worsen the mater-
nal condition.5 To clarify this question, a review of the liter-
ature was conducted to assess maternal and perinatal risks of
expectant management. It is to note that the question of
expectant management in women with HELLP syndrome
will not be addressed in this review.
Randomized Trials
Only two randomized trials have analyzed neonatal morbid-
ity and mortality, and maternal safety, of expectant manage-
ment compared with aggressive management in women with
severe PE remote from term.6,7 Odendaal and coworkers5
conducted a prospective, randomized trial in 58 women be-
tween 28 and 34 weeks’ gestation to establish whether elec-
tive delivery 48 hours after corticosteroid administration (ag-
gressive management) was more beneficial to fetal outcome
than delivery later (expectant management). Twenty (34.5%)
were not randomized because they developed maternal or
fetal complications necessitating delivery during the first 48
hours. Twenty women were randomized to aggressive man-

0146-0005/09/$-see front matter © 2009 Elsevier Inc. All rights reserved. 143
doi:10.1053/j.semperi.2009.02.002
144
agement and 18 to expectant management. All women had
severe PE, defined as blood pressure ⬎180/120 mm Hg on
two occasions at least 30 minutes apart with proteinuria
ⱖ2⫹; blood pressure at 160/110 to 180/120 mm Hg on two
occasions at least 6 hours apart with proteinuria ⱖ2⫹; blood
pressure at 150/100 to 160/110 mm Hg on two occasions at
least 6 hours apart with proteinuria ⱖ3⫹; or blood pressure
ⱖ140/90 mm Hg with proteinuria and “clinical signs of im-
minent eclampsia” (epigastric pain, severe headache, visual
disturbances, nausea, and brisk tendon reflexes). Maternal
indications for delivery were oliguria (⬍400 mL/24 hours),
platelet count ⬍100,000/␮L, abnormal liver function tests,
clinical signs of imminent eclampsia, pulmonary edema, or
attainment of 34 weeks’ gestation. Fetal indications for deliv-
ery were repeated late decelerations or prolonged poor long-
term variability. The authors found that expectant manage-
ment was not associated with increased risk of maternal
complications. In addition, expectant management signifi-
cantly prolonged the gestational age by a mean of 7.1 days
and reduced the rates of neonates requiring ventilation (11%
vs. 35%) or having a complication (33% vs. 75%). However,
this study had limited power to answer this question. Sibai
and coworkers performed a prospective, randomized trial in
which 95 eligible patients with severe PE at 28-32 weeks
were randomly assigned to either aggressive (n ⫽ 46) or
expectant management (n ⫽ 49). At randomization, all
women had severe PE defined by a blood pressure ⱖ160/110
mm Hg associated with a proteinuria ⬎ 500 mg/24 hours and
elevated serum uric acid levels (⬎5 mg/dL). It is of note that
patients with renal disease, insulin-dependent diabetes, con-
nective tissue disease, obstetric complications (such as bleed-
ing, preterm premature rupture of membranes, multifetal
gestation, preterm labor, platelet count ⬍100,000/␮L), or
fetal compromise (such as estimated fetal weight ⬍5th per-
centile or evidence of abnormal fetal testing) were not eligible
in that study. Aggressive management patients were deliv-
ered, either by cesarean or induction, 48 hours after cortico-
steroid prophylaxis. Expectant management patients were
managed with antihypertensives and intensive antenatal fetal
testing. Maternal indications for delivery were uncontrolled
severe hypertension, new onset or persistent severe head-
aches with visual symptoms, epigastric pain, vaginal bleed-
ing, preterm labor, preterm premature rupture of mem-
branes, platelet count ⬍100,000/␮L, fetal distress, or
attainment of 34 weeks’ gestation. Fetal indications for deliv-
ery included presence of repetitive variable or late decelera-
tions, severe oligohydramnios, or biophysical profile persis-
tently ⱕ4. The authors found an average latency period in the
expectant management group of 15.4 days (range, 4-36
days), with a significantly higher gestational age at delivery
(32.9 vs. 30.8 weeks), higher birth weight (1622 vs. 1233 g),
lower incidence of admission to the neonatal intensive care
unit (76% vs. 100%), lower mean days of hospitalization in
the intensive care unit (20.2% vs. 36.6%), and lower inci-
dence of neonatal complications in the expectant manage-
ment group as compared with the aggressive management
group. There was no eclampsia or perinatal death in either
group. In addition, there were no differences in the inci-
B. Haddad and B.M. Sibai
dences of abruptio placentae (4.1% vs. 4.3%). Finally, the
authors concluded that expectant management between 28
and 32 weeks’ gestation in selected women with severe PE,
with frequent monitoring of maternal and fetal status, re-
duces neonatal morbidity. In both studies, 20 (34%) and 18
(16%) women admitted for severe PE were not eligible for
expectant management during the first 48 hours because of
unstable maternal or fetal conditions (see indications for de-
livery in both studies). This highlights the importance of the
selection of patients who may have expectant management.
Finally, these 2 studies have demonstrated an improvement
of neonatal outcome by delaying the delivery 7-15 days with-
out deterioration of maternal condition.6,7
Nonrandomized Trials
Very Early Severe PE
(<25 Weeks’ Gestation)
Studies concerning the management of very early severe PE
are rare. Sibai and coworkers5 reported maternal and perina-
tal outcome of conservative management of 84 consecutive
women with severe PE at earlier gestational age, between 24
and 27 weeks.5 Thirty women were delivered 48 hours after
corticosteroids, and 54 received expectant management.
Treatment included bed rest, antihypertensive therapy to
keep diastolic blood pressure ⬍100 mm Hg, and magnesium
sulfate for 24-72 hours. Among those receiving expectant
management, pregnancy was continued until onset of mater-
nal and/or fetal complications. The average length of preg-
nancy prolongation in the expectant management group was
13 days (range, 2-26 days). Patients having expectant man-
agement had significantly higher perinatal survival rates and
lower neonatal morbidities. In addition, maternal complica-
tions were minimal. Pattinson and coworkers8 reported ma-
ternal and perinatal outcomes in 45 women with severe PE
before 28 weeks’ gestation that had conservative manage-
ment. Treatment included bed rest, antihypertensive treat-
ment to keep diastolic blood pressure between 90 and 100
mm Hg, betamethasone administration after 26 weeks’ ges-
tation, and intensive fetal and maternal monitoring. The ad-
mission-to-delivery interval averaged 14 days. Perinatal mor-
tality rates were 100% and 62% when women with severe PE
were expectantly managed at ⬍24 weeks and 24-28 weeks,
respectively. In a recent retrospective study, Gaugler-Senden
and colleagues9 evaluated pregnancy outcome in 26 preg-
nancies with an early onset of PE before 24 weeks’ gestation,
2 of them being twins. The median prolongation of the preg-
nancy was 24 days (range, 3-46 days). The overall perinatal
mortality was 82%: 19 fetal deaths and 4 neonatal deaths.
Major maternal complications were observed in 65% of
women analyzed: maternal death in 1 case, eclampsia in 5
cases, HELLP syndrome in 16 cases, and pulmonary edema
in 4 cases. Because of poor perinatal outcome and high major
maternal morbidity, the authors concluded that expectant
management should not be considered as a routine treatment
option in these patients. Budden and coworkers10 analyzed in
a retrospective study the usefulness of expectant manage-
Expectant management in pregnancies with severe PE
ment in 31 women with PE admitted at ⬍25 weeks’ gesta-
tion, all of them with singleton pregnancy. Of the 14 women
admitted ⬍23 weeks, no babies survived. Of the 17 women
admitted at 24-246/7 weeks, 9 had a baby discharged alive.
Neurodevelopmental outcome assessed at 18 months in 8 of
the 9 survivors showed moderate to mild disability in 4 of
them (50%). Finally, only 25% of the babies born from moth-
ers admitted at 24-246/7 weeks and expectantly managed
were free from neurological disability assessed at 18 months.
On the other hand, the rates of maternal complications were
extremely high; HELLP syndrome was observed in 20 (65%)
women, renal impairment in 8 (26%) patients, and a stay in
intensive care unit for more than 48 hours in 12 (39%) pa-
tients. Sezik and coworkers11 reported expectant manage-
ment in 55 patients with severe PE at ⬍25 weeks. There were
52 still births and 1 neonatal death with perinatal mortality of
96.3%. More recently, Bombrys and coworkers12 reported a
study where 46 patients (51 fetuses) with severe PE at less
than 27 weeks were expectantly managed. Of them, 20 pa-
tients (24 fetuses) were admitted at less than 25 weeks’ ges-
tation. Corticosteroids were administered beyond 23 weeks.
From the 17 babies born of those admitted ⬍24 weeks’ ges-
tation, only 2 (12%) were discharged alive. For those admit-
ted at 24-246/7, the perinatal survival rates were 5 of 7 (71%).
Expectant management in those women resulted in a high
rate of maternal complication, with 11 (55%) women having
at least one major complication, particularly the 8 (40%) with
HELLP syndrome. These studies suggest that pregnancy ter-
mination, rather than expectant management, should be se-
riously considered in women with severe PE at ⬍24 weeks’
gestation.2,8-12
Early Severe PE
(at 24-34 Weeks’ Gestation)
Odendaal and coworkers13 reported maternal and perinatal
outcome in 129 patients with severe PE at ⬍34 weeks. The
admission-to-delivery interval was 11 days. None of the pa-
tients died or developed eclampsia. The perinatal mortality
rate was 22%. An abruptio placenta was the cause of fetal
death in 36% of the 14 fetuses that died in utero. Olah and
coworkers14 retrospectively compared conservative manage-
ment (n ⫽ 28) with stabilization and early intervention (n ⫽
28) in women with severe PE between 24 and 32 weeks.
Those women managed conservatively gained a mean of 9.5
days. There were fewer babies with 1 or more neonatal com-
plications in those cases managed conservatively (28.6%) as
compared with those in the early intervention group
(64.3%). All women in the early intervention group recov-
ered with no severe complications. In contrast, those women
managed conservatively had a higher incidence of HELLP (2
cases) and ELLP syndrome (2 cases). In 1 case, temporary
renal dialysis was required. Visser and Wallenburg15 re-
ported maternal and perinatal outcomes of expectant man-
agement of 254 consecutive patients with severe PE between
20 and 32 weeks’ gestation. Management included bed rest,
pharmacologic vasodilatation by dihydralazine and plasma
volume expansion under central hemodynamic monitoring,
145
and occasional use of diazepam in women with clinical signs
of imminent eclampsia. Corticosteroid prophylaxis was not
part of the management at that time. It is of note that women
who had eclampsia (4.7%) or HELLP syndrome (33.5%) be-
fore admission were not excluded from expectant manage-
ment. The admission-to-delivery interval average was 14
days, with a termination within 48 hours in 12.6%. The
perinatal mortality rate was 20.5%. None of the women died,
and only 1 woman had eclampsia. HELLP syndrome was
observed in 8%, and abruptio placentae in 5%. Later, Visser
and Wallenburg16 compared, in women with severe PE at
⬍34 weeks, maternal and perinatal outcomes of pregnancies
in those with (n ⫽ 128) and without HELLP syndrome (n ⫽
128).16 Both groups were treated with plasma volume expan-
sion and pharmacologic vasodilatation under central hemo-
dynamic monitoring with the aim of prolonging gestation
and enhancing fetal maturity. Corticosteroids to accelerate
fetal lung maturity and magnesium sulfate to prevent
eclampsia were not used in the study period. The admission-
to-delivery interval was 10 days in HELLP patients and 14
days in patients without HELLP. Perinatal and neonatal mor-
talities were 14.1% and 5.9% in patients with HELLP and
14.8% and 6.8% in patients without HELLP, respectively.
Eclampsia occurred in 1.6% women with HELLP and in
0.8% women without HELLP.
Hall and coworkers17,18 evaluated, in a prospective study,
maternal and perinatal outcomes in 340 women with severe
PE expectantly managed between 24 and 34 weeks. Up to 3
oral antihypertensives (methyldopa, prazosin, and nifedi-
pine) were used to control blood pressure to levels ⬍160/110
mm Hg. Oral nifedipine and intravenous dihydralazine were
used to control hypertensive peaks. Magnesium sulfate was
given to prevent recurrence of eclampsia, whereas prophy-
laxis with this drug was used during intrapartum manage-
ment, according to the discretion of the physicians. A com-
plete course of betamethasone was given to women at the end
of the 27th week of gestation, and thereafter, 12 mg were
given weekly until 33 weeks’ gestation or delivery. Fetal sur-
veillance included 6-hourly heart rate monitoring, and
weekly Doppler and ultrasound evaluation of the fetus every
2 weeks. Failure to control blood pressure or development of
severe maternal or fetal complications, or reaching 34 weeks’
gestation, was an indication for delivery. Fetal viability was
set at 28 weeks’ gestation with a minimum weight of 800 g.
The mean admission-to-delivery interval was 11 days. The
perinatal mortality rate was 24 per 1000 (ⱖ 1000 g/7 days) or
44.1 per 1000 (ⱖ500 g/7 days) with a neonatal survival rate
of 94%. Three pregnancies (0.8%) were terminated before
viability, and only 2 (0.5%) intrauterine deaths occurred,
both because of placental abruption. Neonatal intensive care
was necessary in 40.7% of cases. No maternal deaths oc-
curred during expectant management. Twenty-seven percent
of women experienced a major complication, but few had
poor outcomes. It is of note, however, that 20% of women
had placentae abruptio and 1.2% had eclampsia. One woman
required dialysis, and only 3 (0.8%) women required admis-
sion to the intensive care unit. Vigil-De Gracia and cowork-
ers19 aimed to compare maternal and perinatal outcome in
146
women expectantly managed with severe PE (n ⫽ 100) to
those with superimposed PE to chronic hypertension (n ⫽
29). For both groups, management included bed rest, treat-
ment with bolus of hydralazine or labetalol, or oral nifedipine
to control severe hypertension (⬎160/110 mm Hg), plasma
volume expansion (900 mL Ringer’s lactate mixed with 100
mL 25% albumin), magnesium sulfate to prevent eclampsia,
and corticosteroids to enhance fetal lung maturity. Indica-
tions for delivery during expectant management were severe
uncontrolled hypertension, HELLP syndrome, abruptio pla-
centae, persistent headaches or visual disturbances, fetal dis-
tress, and attainment of 34 weeks’ gestation. Women with
severe PE had an average prolongation of 8.4 days. Expectant
management was associated with HELLP syndrome and ab-
ruptio placentae in 9% of the cases. None of the women had
eclampsia or disseminated intravascular coagulopathy. Peri-
natal deaths and respiratory distress syndrome were ob-
served in 7% and 18%, respectively, in women with severe
PE expectantly managed. Haddad and coworkers20 under-
took a prospective observational study to determine maternal
and perinatal outcomes after expectant management of se-
vere PE between 24 and 33 weeks’ gestation. Of the 381
women admitted for severe PE, 142 (37%) were delivered
during the first 48 hours after admission for maternal rea-
sons, mainly eclampsia (8%), antepartum HELLP syndrome
(34%), abruptio placentae (13%), or fetal abnormal fetal
heart rate or severe oligohydramnios. Finally, 239 (63%) re-
mained undelivered after antenatal steroid prophylaxis was
performed. These women were eligible for expectant man-
agement. Maternal indications for delivery during expectant
management were major maternal complications (eclampsia,
HELLP syndrome, abruptio placentae, disseminated intra-
vascular coagulopathy, pulmonary edema, and acute renal
failure), except for maternal death, severe uncontrolled hy-
pertension, despite maximum doses of combined antihyper-
tensive therapy (nicardipine and labetalol), persistent head-
aches or visual disturbances, persistent epigastric pain, low
platelet count (⬍100,000 cells/␮L), and oliguria. Fetal indi-
cations for delivery during expectant management were ab-
normal fetal heart rate monitoring (repeated late decelera-
tions or decreased long-term variability), severe IUGR, and
oligohydramnios. Pregnancy prolongation and maternal and
perinatal morbidities were analyzed according to the gesta-
tional age at the time of expectant management: 24-28, 29-
31, and 32-33 weeks. The days of pregnancy prolongation
were significantly higher among those managed at ⬍29
weeks (6 days) compared with the other groups (4 days).
There were 13 perinatal deaths (5.4%): 12 in those managed
at ⬍29 weeks [median gestational age at delivery: 26.2 weeks
(24.8-30.6 weeks), median birth weight: 650 g (360-960 g)]
and 1 in those managed at 29-31 weeks. Neonatal morbidi-
ties were significantly higher among those managed at ⬍29
weeks compared with the other groups. There were no in-
stances of maternal death or eclampsia. Maternal morbidities
were similar among the groups.
Oettle and coworkerss21 analyzed in women with early
severe PE the feasibility of expectant management in a sec-
ondary care center in close cooperation with the tertiary cen-
B. Haddad and B.M. Sibai
ter. They included 131 women between 24 and 34 weeks’
gestation with severe PE. Women were delivered on achiev-
ing 34 weeks or if fetal distress or major maternal complica-
tions developed. Transfer to the tertiary center was individ-
ualized. Interestingly, 88.5% of the women were managed
entirely at the secondary hospital. Major maternal complica-
tions occurred in 33.6% of the cases, with placental abrup-
tion in 23% of the women. Two women required intensive
care admission, and one maternal death occurred. A mean of
11.6 days was gained before delivery, with the mean delivery
gestation being 31.8 weeks. There were four intrauterine
deaths. The perinatal mortality rate (ⱖ1000 g) was 44.4/
1000, and the early neonatal mortality rate (ⱖ500 g) was
30.5/1000. The authors finally found that maternal and peri-
natal outcomes were comparable to those achieved by other
tertiary units.
Recently, Bombrys and coworkers22 reported expected
management in 66 patients with severe PE between 270/7 and
336/7 weeks; 5 of the patients had twin gestation. Median time
of pregnancy prolongation was 5 days (range, 3-35 days).
Among the 71 fetuses, there were no fetal deaths and only 1
neonatal death in an infant born at 27 weeks for a perinatal
survival rate of 98.6%. Seven of the 66 (11%) had abruptio
placentae, 5 (8%) had HELLP syndrome, 6 (9%) developed
pulmonary edema, and 2 (3%) had renal insufficiency. Over-
all, 18 (27%) had one or more of these complications. Most of
the studies found an admission-to-delivery interval of 10-14
days.2 This is of importance because every week gained in
women at ⬍32 weeks may improve neonatal outcome.2,5 In
addition, randomized studies did not show any increase of
maternal morbidity during expectant management.6,7
Specific Management
Endothelial cell injury is thought to be one of the mecha-
nisms that lead to PE. Endothelial cells of several organs, such
as the brain, kidneys, liver, or placenta, may be involved. As
a result, the severity of the disease will depend on the degree
of organ involvement. Even in severe cases, some women will
have a stabilization of the disease and would be eligible for
expectant management. Conversely, some others will have a
rapid deterioration of the maternal or fetal conditions that
will necessitate expeditious delivery, regardless of gestational
age. In most of the trials, women with complications, such
as eclampsia, abruption, HELLP syndrome, platelet count
⬍100,000/␮L, or fetal growth retardation, were not consid-
ered eligible for expectant management.
Fetal Growth Retardation
Some studies have recently questioned the possibility to pro-
long gestation in women with fetal growth retardation.
Chammas and coworkers23 aimed to determine, in an obser-
vational study, the frequency of fetal deterioration with ex-
pectant management of PE at ⬍34 weeks’ gestation and to
evaluate whether the presence of intrauterine growth restric-
tion on admission was associated with a shorter admission-
to-delivery interval or more deliveries resulting from non-
Expectant management in pregnancies with severe PE
reassuring fetal status in comparison with pregnancies with
PE but without intrauterine growth restriction. Admission-
to-delivery interval was found significantly shorter in women
with severe PE associated with intrauterine growth restriction
when compared with those without intrauterine growth re-
striction (3.1 days vs. 6.6 days, respectively). Eighty-six per-
cent and 57% of fetuses with intrauterine growth restriction
were delivered before 1 week and for fetal indications, re-
spectively. Finally, neonatal outcomes, as reflected by
APGAR scores, number of admissions to and duration of stay
in the neonatal intensive care unit, and neonatal mortality
rates were similar. Shear and coworkers24 studied 155 single-
ton pregnancies with severe PE at ⬍34 weeks’ gestation over
a 10-year period. Outcomes of both mother and fetus were
stratified according to gestational age and the severity of fetal
growth restriction ⱕ3rd percentile, 4th to 5th percentile,
⬎5th to 10th percentile, and ⬎10th percentile. Perinatal
mortality rate was 3.9%. The authors found that gestational
age of ⬍30 weeks’ gestation was the strongest variable that
affected perinatal outcome, whereas fetal growth restriction
played a marginal role. They concluded that expectant man-
agement should strongly be recommended in fetuses at ⬍30
weeks, irrespective of fetal growth restriction, and delivery
should be considered at ⬍30 weeks’ gestation. In contrast, a
recent study by Haddad and coworkers25 revealed that ex-
pectant management was associated with increased rate of
fetal deaths in those delivering a fetal growth-retarded infant.
The Use of Umbilical
Artery Doppler Velocimetry
The use of Doppler flow velocity waveform in the manage-
ment of high-risk pregnancies can reduce perinatal mortality.
In a prospective, observational, multicenter study, Karsdorp
and coworkers26 evaluated perinatal outcome in 3 groups of
pregnancies according to umbilical artery Doppler status:
those with positive end diastolic velocities (n ⫽ 214), absent
end diastolic velocities (n ⫽ 178), and reversed end diastolic
velocities (n ⫽ 67). The overall perinatal mortality rate was
28%. The perinatal mortality rates were significantly different
in women with positive end diastolic flow (4%) when com-
pared with those of absent end diastolic flow (41%) and
reversed end diastolic flow (75%). The odds for perinatal
mortality were 4.0 and 10.6 in pregnancies complicated by
absent end diastolic flow and reversed end diastolic flow,
respectively, even after adjustment for menstrual age. Yoon
and coworkers27 analyzed the relationship between abnor-
mal umbilical artery Doppler waveform and adverse perinatal
outcome in 72 consecutive patients with PE. Among them,
42 had severe PE. Thirty-seven women were found to have an
abnormal umbilical artery Doppler waveform; of these, 15
had absent or reversed end diastolic flow. Patients with ab-
normal umbilical artery velocimetry were found to have a
significantly higher rate of complications, including cesarean
section for fetal distress, preterm delivery, low APGAR
scores, significant neonatal morbidity, and perinatal death,
than did patients with a normal waveform. In addition, logis-
tic regression analysis indicated that an abnormal umbilical
147
artery waveform was a significant independent predictor for
the development of adverse perinatal outcome (odds ratio
14.2). Specific analysis of the 13 perinatal death showed that
12 of them occurred in women with absent or reversed end
diastolic flow. Finally, Pattinson and coworkers28 analyzed
the usefulness of umbilical arterial Doppler velocimetry in
the management of high-risk pregnancies at ⬎28 weeks’ ges-
tation. Three groups of patients were analyzed: women with
absent end diastolic flow (n ⫽ 20), women with IUGR and
positive end diastolic flow (n ⫽ 103), and women with hy-
pertensive disease and positive end diastolic flow (n ⫽ 89).
Women of each group were randomized to have the results of
Doppler velocimetry revealed or unrevealed to the physician.
In the study management, women at ⬎30 weeks’ gestation or
estimated fetal weight ⬎1000 g with absent or reversed end
diastolic flow revealed to the physician were delivered. Over-
all, perinatal mortality was similar in both groups. In women
with absent end diastolic flow, however, perinatal mortality
was significantly decreased in those women whose physi-
cians knew the result (1 of 10) compared with those in whom
the result remained blinded (6 of 10).
Finally, absent or reversed end diastolic flow may be a
marker for pregnancy termination in women with severe PE,
particularly beyond 32 weeks’ gestation.
Computerized
Antepartum Cardiotocography
Antenatal fetal heart rate monitoring is a major part of fetal
assessment. The introduction of computerized antepartum
measurements during the last decade has certainly contrib-
uted to improve the assessment of fetal well-being. Guzman
and coworkers29 determined the efficacy of individual fetal
heart rate indexes, as determined by computer analysis of the
fetal heart rate tracing, in detecting fetal acidemia at birth
(umbilical artery pH ⬍7.20) in growth-restricted fetuses.
Thirty-eight growth-restricted fetuses at 26-37 weeks’ gesta-
tion from pregnancies with abnormal uterine and/or umbil-
ical artery Doppler velocimetry were analyzed. Seventy-four
percent of the pregnancies were complicated by some forms
of hypertensive disease. On linear regression, the duration of
episodes of low variation in minutes (r ⫽ 0.77, r2 ⫽ 0.59),
short-term (r ⫽ 0.72, r2 ⫽ 0.52) and long-term (r ⫽ 0.69,
r2 ⫽ 0.47) variation in milliseconds (ms) were significantly
related to umbilical artery pH at birth. The optimal cut-offs
for these three parameters were ⬍45 minutes, ⬎3.5 ms, and
⬍15 ms, respectively. Recently, Serra and coworkers30 aimed
to assess, in a retrospective study, the clinical value of the
short-term fetal heart rate variation (STV) for timing the de-
livery of severely growth-retarded fetuses. A total of 257 fe-
tuses with a birthweight ⬍3rd percentile and a last comput-
erized cardiotocography performed within 24 hours of
delivery were analyzed. Of them, 49% had PE. There were no
stillbirths or neonatal deaths (NNDs) within 24 hours in the
study population. The authors found a correlation between
decreasing STV and earlier deliveries (P ⬍ 0.001), lower
birthweight (P ⬍ 0.001), lower umbilical artery pH at birth
(P ⬍ 0.001), worse acid– base status at birth (P ⬍ 0.001), and
148
worse postnatal outcome (P ⬍ 0.002). The STV was able to
predict the presence or absence of metabolic acidemia (area
under the receiver operating characteristic curve 0.75, P ⬍
0.001). The risk of metabolic acidemia increased as the an-
tepartum STV decreased, with an optimal cut-off level of
ⱕ3.0 ms (positive and negative predictive values 64.6% and
86.6%, respectively). An STV ⱕ3.0 ms was associated with
higher rate of metabolic acidemia and early NNDs compared
with an STV ⬎3.0 ms (54.2 vs. 10.5% and 8.3 vs. 0.5%,
respectively; P ⬍ 0.001). It is to note that deaths that oc-
curred in the former group were all due to extreme prema-
turity and very low birthweight. Finally, an STV cut-off level
of ⱕ3 ms seems to be an important marker of adverse peri-
natal outcome in severely growth-retarded fetuses.
Corticosteroid Prophylaxis
Because of the stress of PE, respiratory distress syndrome has
been thought to be decreased in the beginning of the past
decade. Schiff and coworkers3 conducted a matched cohort
study to determine whether there is an increased incidence of
pulmonary maturity in premature fetuses of pre-eclamptic
women compared with fetuses of matched controls. A total of
127 pre-eclamptic women who had undergone amniocente-
sis for pulmonary maturity assessment were matched to non-
hypertensive women with preterm labor who had undergone
the same procedure. The authors found no difference in the
incidence of an immature result between the pre-eclamptic
and matched control groups (39.4% vs. 38.6%). In addition,
women with mild (n ⫽ 63) and severe (n ⫽ 64) PE showed
no differences in comparison with their matched controls.
This study excluded the beneficial effect of PE on fetal lung
maturity. Recently, Amorim and coworkers31 analyzed the
efficacy of corticosteroid therapy in the prevention of respi-
ratory distress syndrome in women with severe PE. The au-
thors performed a prospective, double-blind, randomized
trial in which 218 women were enrolled at 26-34 weeks. A
total of 110 women received betamethasone (12 mg/day for 2
days administered intramuscularly, and then 12 mg/day re-
peated once a week) and 108 received placebos. In this study,
the frequency of respiratory distress syndrome was signifi-
cantly lower in the corticosteroid group as compared with the
placebo group [23% vs. 43%, respectively; relative risk (RR):
0.53, 95% confidence interval (CI): 0.35-0.82]. In addition,
relative risks of intraventricular hemorrhage, patent ductus
arteriosus, and perinatal infection were significantly de-
creased in the corticosteroid group: 0.35 (95% CI: 0.15-
0.86), 0.27 (95% CI: 0.08-0.95), and 0.39 (95% CI: 0.39-
0.97), respectively. Neonatal mortality rate was lower in the
corticosteroid group as compared with the placebo group
(14% vs. 28%, respectively; RR: 0.5, 95% CI: 0.28-0.89).
There was no significant difference in the frequency of still-
birth between the 2 groups. There was increased risk of ges-
tational diabetes but of no other maternal complication after
corticosteroid therapy. In addition, mean blood pressures
were similar in both groups. It is now recommended that
women with severe PE remote from term should have corti-
costeroid therapy to enhance fetal lung maturity.
B. Haddad and B.M. Sibai
Proteinuria
Schiff and coworkers32 and Hall and coworkers33 found that
the amount of proteinuria during expectant management of
severe PE did not correlate with admission-to-delivery inter-
val or maternal or perinatal outcomes. In addition, these
authors found no difference in maternal–perinatal outcome
between those with proteinuria ⱖ5 g/24 hours or ⬍5 g/24
hours. In addition, Chua and Redman reported no residual
renal dysfunction in women who had expectant management
and proteinuria ⬎5 g/day.34 Recently, Newman and col-
leagues35 compared maternal or neonatal morbidity in pre-
eclamptic women having mild (⬍5 g/24 hours, n ⫽ 125),
severe (5-9.9 g/24 hours, n ⫽ 43), or massive (⬎10 g/24
hours, n ⫽ 41) proteinuria.35 Patients with underlying renal
disease were excluded. No significant differences in maternal
morbidity were seen. Massive proteinuria was associated
with earlier onset of PE, earlier gestational age at delivery, and
higher rates of prematurity complications. After correction
for prematurity, massive proteinuria had no significant effect
on neonatal outcomes. Therefore, it is not our practice to
deliver women with severe PE solely because of proteinuria
⬎5 g/day.
Plasma Volume Expansion
There are some doubts about the usefulness of plasma vol-
ume expansion in women with severe PE. Only one well-
designed study addressed this issue. Ganzevoort and co-
workers analyzed, in a prospective randomized trial, 216
patients at 24-34 weeks’ gestation with severe PE, HELLP
syndrome, or severe fetal growth restriction with pregnancy-
induced hypertension.36 A total of 111 patients were ran-
domly allocated to plasma volume expansion (250 mL hy-
droxyethyl starch 6% given twice daily over 4 hours) and 105
to the control group (intravenous fluid restriction). The au-
thors found a trend toward less prolongation of pregnancy
(median 7.4 days vs. 11.5 days; P ⫽ 0.054) and more infants
requiring oxygen treatment ⬎21% (66 vs. 46; P ⫽ 0.09) in
the treatment group. In addition, no differences were ob-
Table 1 Maternal Indications for Expeditious Delivery Within
48 Hours in Severe Pre-eclamptic Women
Uncontrolled severe hypertension (>160 mm Hg systolic
or >110 mm Hg diastolic) despite maximum doses
recommended of at least 2 antihypertensive agents
Eclampsia
Pulmonary edema
Abruptio placentae
Oliguria (<0.5 mL/kg/hour) that does not resolve with fluid
intake
Signs of imminent eclampsia (persistent severe headache
or visual disturbance)
Persistent epigastric pain or right upper quadrant
tenderness
Rapid deterioration of HELLP syndrome or platelet counts
<100,000/␮L
Deterioration of renal function (serum creatinine >1.4
mg/dL)
Expectant management in pregnancies with severe PE
Table 2 Fetal Indications for Expeditious Delivery Within 48
Hours in Severe Pre-eclamptic Women
Repetitive late decelerations
Severe variable decelerations
Short-term variability <3 ms
Biophysical profile <4 on two occasions at 4 hours apart
Severe oligohydramnios
served between the 2 groups in perinatal outcomes or in
perinatal deaths. There was no difference in major maternal
morbidity (total 11%), but there were more cesarean sections
in the treatment group (98% vs. 90%; P ⬍ 0.05). The authors
concluded that plasma volume expansion in expectant man-
agement of women with severe hypertensive disease did not
improve maternal or fetal outcome.
Candidates for
Expectant Management
Regarding maternal and fetal features, we have adopted in
our practice additional criteria for expeditious delivery
within the 48 hours (Tables 1 and 2) in women with severe
PE. This delay allows a full course of corticosteroid treatment.
We believe that the risks of delay beyond 48 hours outweigh
the benefits. Certainly, some of these indications, such as
rapid deterioration of HELLP syndrome, abruptio placentae,
eclampsia, and fetal distress, may require delivery before 48
hours. Conversely, other indications, such as severe blood
pressure controlled by antihypertensives during expectant
management, may be an opportunity to prolong pregnancy
beyond 48 hours. When using these criteria in our study, we
found, along with others, that 63% (239/381) of women with
severe PE are eligible for expectant management beyond the
first 48 hours after admission.20
Our Guidelines for
Expectant Management
After admission, patients are observed in the labor and deliv-
ery unit depending on severity of clinical and biological re-
sults (Fig. 1). Betamethasone to enhance fetal lung maturity is
immediately started between 24 and 34 weeks= gestation.
Antihypertensive drugs are administered to keep systolic
blood pressure at 130-150 mm Hg and diastolic blood pres-
sure at 80-100 mm Hg. We use oral nifedipine or labetalol as
first-line treatment and intravenous bolus of nicardipine or
labetalol when systolic blood pressure is ⱖ160 mm Hg
or diastolic blood pressure ⱖ110 mm Hg. Hemoglobin
with platelets, serum levels of creatinine, aspartate amino
transferase, lactate dehydrogenase, bilirubin, fibrinogen, D-
dimers, prothrombin and partial thromboplastin times, and
24-hour urine collection for total proteinuria are obtained.
During the observation period, intravenous Ringer’s lactate
solution with 5% dextrose is administered at 100 mL/h. Fetal
condition at entry is assessed by fetal biophysical profile,
specific analysis of fetal heart rate monitoring, the estimation
149
of fetal growth and amniotic fluid volume, and the study of
umbilical artery Doppler waveform. If the patient and her
fetus have stable status, and are judged suitable for expectant
management by the conditions enumerated in Tables 1 and
2, then she is treated and observed in the antepartum unit.
Maternal surveillance includes blood pressure control and
the search for signs of imminent eclampsia (persistent severe
headaches or visual disturbance) every 4 hours, hemoglobin,
platelet count, liver enzymes, creatinine, performed daily or
every other day, depending on clinical symptoms and labo-
ratory findings. Twenty-four-hour urine proteinuria is per-
formed at entry and is not analyzed again once severe PE has
been documented.32,33 Twenty-four-hour urine volume and
maternal weight are assessed every day. In case of gestational
age at 32 weeks’ gestation or more and estimation of fetal
growth below fifth percentile or reversed end diastolic flow,
the pregnancy is terminated after corticosteroid prophylaxis.
Fetal surveillance during expectant management includes
fetal heart rate monitoring at least daily. Chari and cowork-
ers37 found that daily antenatal testing in pregnancy with
severe PE reduced the occurrence of stillbirth or fetal com-
promise at delivery. We perform weekly ultrasound exami-
nation to assess amniotic fluid volume and umbilical artery
Doppler ultrasound. In pregnancies with IUGR, we perform
antepartum fetal heart rate monitoring twice daily and um-
bilical artery Doppler waveform twice weekly. During that
period, the patient remains at the hospital. In the US, mag-
nesium sulfate for seizure prophylaxis is administered at en-
try, and if the patient is judged suitable for expectant man-
agement, magnesium sulfate is discontinued. In contrast, in
France, magnesium sulfate is started only in women in whom
expeditious delivery is decided. Although rapid progression
of severe PE has not been considered an indication for expe-
ditious delivery (Table 1), the obstetrician should rely on
his/her clinical judgment to anticipate timing of delivery.
Severe organ involvement or pre-existing medical condi-
tions predisposing to severe maternal or perinatal outcomes,
such as insulin-dependent diabetes, chronic hypertension, or
lupus, should be taken into account in the decision of expect-
ant management. Gestational age limits of expectant manage-
ment depend on the availability of neonatal intensive care
facilities, the neonatal complications at a given gestational age
in that population and in the institution, and the feasibility of
maternal and fetal surveillance. Because of worst perinatal
outcome at 25-27 weeks’ gestation, these women should not
be expectantly managed like those at 31-33 weeks’ gestation.
In fact, several studies showed that admission-to-delivery
interval is higher at earlier gestation ages,17,18,20 mainly be-
cause obstetricians are reluctant to take inconsiderable risks
for the mother above 31 weeks’ gestation. If preterm labor
supervenes, no specific tocolysis is performed. Magnesium
sulfate is started and vaginal route is accepted if there are no
contraindications. At any time during expectant manage-
ment, the development of any sign described in Tables 1 and
2 necessitates delivery. Delivery generally occurs by cesarean
section. Some teams, however, recommend induction of la-
bor in these situations. The reported rates of vaginal delivery
are very low: 18.5%17,18 to 27%.7 In our study, only 5% of
150
Figure 1 Algorithm for management of women with severe preeclampsia remote from term.
women had vaginal delivery.20 It is our policy that these
patients with an unripe cervix at ⱕ30 weeks be delivered by
cesarean section because prolonged labor may be detrimental
for either the mother or the fetus, and the success rate is
usually low.38,39 Finally, we believe that women with severe
PE reaching 34 weeks’ gestation should not have expectant
management beyond 34 weeks because maternal risks out-
weigh perinatal benefits. Moreover, expectant management
of selected severe PE remote from term should, whenever
possible, be realized in a tertiary care center. A detailed algo-
rithm for management of these patients is described in Figure 1.
Conclusions
The main objective of expectant management in women with
severe PE remote from term is to improve neonatal outcome.
Maternal conditions, however, may worsen during expectant
management. This highlights the importance of balancing the
risks between maternal and perinatal outcomes. Tradition-
ally, women with severe PE remote from term are delivered
expeditiously, regardless of gestational age. We here have
B. Haddad and B.M. Sibai
reported several retrospective, case-control, observational,
prospective, or randomized trials in which expectant man-
agement in women with severe PE was feasible in well-se-
lected patients without prejudicing maternal safety, and we
have described our rationale and guidelines for this manage-
ment. The first 48 hours after admission are critical because
corticosteroid to enhance fetal lung maturity will be started
and women eligible to expectant management will be se-
lected during that period. This management should be per-
formed in a tertiary care center.
References
1. American College of Obstetricians and Gynecologists: Hypertension in
Pregnancy. Washington: The College. Technical Bulletin No. 219,
1996
2. Sibai BM, Barton JR: Expectant management of severe preeclampsia
remote from term: Patient selection, treatment, and delivery indica-
tions. Am J Obstet Gynecol 196:514.e1-514.e9, 2007
3. Schiff E, Friedman SA, Mercer BM, et al: Fetal lung maturity is not
accelerated in preeclamptic pregnancies. Am J Obstet Gynecol 169:
1096-1101, 1993
4. Crowley P: Antenatal corticosteroid therapy: A metal-analysis of the
Expectant management in pregnancies with severe PE
randomized trials, 1972-1994. Am J Obstet Gynecol 173:322-335,
1995
5. Sibai BM, Akl S, Fairlie F, et al: A protocol for managing severe pre-
eclampsia in the second trimester. Am J Obstet Gynecol 163:733-738,
1990
6. Odendaal HJ, Pattinson RC, Bam R, et al: Aggressive or expectant man-
agement for patients with severe preeclampsia between 28-34 weeks’
gestation: A randomized controlled trial. Obstet Gynecol 76:1070-
1075, 1990
7. Sibai BM, Mercer BM, Schiff E, et al: Aggressive versus expectant man-
agement of severe preeclampsia at 28-32 weeks’ gestation: A random-
ized controlled trial. Am J Obstet Gynecol 171:818-822, 1994
8. Pattinson RC, Odendaal HJ, du Toit R: Conservative management of
severe proteinuric hypertension before 28 weeks’ gestation. S Afr Med
J 73:516-518, 1988
9. Gaugler-Senden IP, Huijssoon AG, Visser W, et al: Maternal and peri-
natal outcome of preeclampsia with an onset before 24 weeks’ gesta-
tion. Audit in a tertiary referral center. Eur J Obstet Gynecol Reprod
Biol 128:216-221, 2006
10. Budden A, Wilkinson L, Buksh MJ, et al: Pregnancy outcome in women
presenting with pre-eclampsia at less than 25 weeks gestation. Aust
N Z J Obstet Gynaecol 46:407-412, 2006
11. Sezik M, Tuncay G, Yapar EG: Prediction of adverse neonatal outcomes
in preeclampsia by absent or reversed end-diastolic flow velocity in the
umbilical artery. Gynecol Obstet Invest 57:109-113, 2004
12. Bombrys AE, Barton JR, Nowacki EA, et al: Expectant management of
severe preeclampsia at less than 27 weeks’ gestation: Maternal and
perinatal outcomes according to gestational age by weeks at onset of
expectant management. Am J Obstet Gynecol 199:e1-e6, 2008
13. Odendaal HJ, Pattinson RC, du Toit R: Fetal and neonatal outcome in
patients with severe pre-eclampsia before 34 weeks. S Afr Med J 71:
555-558, 1987
14. Olah KS, Redman CW, Gee H: Management of severe, early pre-
eclampsia: Is conservative management justified? Eur J Obstet Gynecol
Reprod Biol 51:175-180, 1993
15. Visser W, Wallenburg HC: Maternal and perinatal outcome of tempo-
rizing management in 254 consecutive patients with severe pre-
eclampsia remote from term. Eur J Obstet Gynecol Reprod Biol 63:147-
154, 1995
16. Visser W, Wallenburg HC: Temporising management of severe pre-
eclampsia with and without the HELLP syndrome. Br J Obstet Gynae-
col 102:111-117, 1995
17. Hall DR, Odendaal HJ, Steyn DW, et al: Expectant management of early
onset, severe pre-eclampsia: Maternal outcome. Br J Obstet Gynaecol
107:1252-1257, 2000
18. Hall DR, Odendaal HJ, Kirsten GF, et al: Expectant management of
early onset, severe pre-eclampsia: Perinatal outcome. Br J Obstet
Gynaecol 107:1258-1264, 2000
19. Vigil-De Gracia P, Montufar-Rueda C, Ruiz J: Expectant management of
severe preeclampsia and preeclampsia superimposed on chronic hy-
pertension between 24 and 34 weeks’ gestation. Eur J Obstet Gynecol
Reprod Biol 107:24-27, 2003
20. Haddad B, Deis S, Goffinet F, et al: Maternal and perinatal outcomes
during expectant management of 239 severe preeclamptic women be-
tween 24 and 33 weeks’ gestation. Am J Obstet Gynecol 190:1590-1597,
2004
21. Oettle C, Hall D, Roux A, et al: Early onset severe pre-eclampsia: Ex-
151
pectant management at a secondary hospital in close association with a
tertiary institution. Br J Obstet Gynaecol 112:84-88, 2005
22. Bombrys AE, Barton JR, Habli M, et al: Expectant management of severe
preeclampsia at 270/7-336/7 weeks’ gestation: Maternal and perinatal
outcomes according to gestational age by X weeks at onset of expectant
management. Am J Perinatol 2009 (in press)
23. Chammas MF, Nguyen TM, Li MA, et al: Expectant management of
severe preterm preeclampsia: Is intrauterine growth restriction an in-
dication for immediate delivery? Am J Obstet Gynecol 183:853-858,
2000
24. Shear RM, Rinfret D, Leduc L: Should we offer expectant management
in cases of severe preterm preeclampsia with fetal growth restriction?
Am J Obstet Gynecol 192:1119-1125, 2005
25. Haddad B, Kayem G, Sies S, et al: Are perinatal and maternal outcomes
different during expectant management of severe preeclampsia in the
presence of intrauterine growth restriction? Am J Obstet Gynecol 196:
237.e1-237.e5, 2007
26. Karsdorp VHM, van Vugt JMG, van Geijn HP, et al: Clinical significance
of absent or reversed end diastolic velocity waveforms in umbilical
artery. Lancet 344:1664-1668, 1994
27. Yoon BH, Lee CM, Kim SW: An abnormal umbilical artery waveform: A
strong and independent predictor of adverse perinatal outcome in pa-
tients with preeclampsia. Am J Obstet Gynecol 171:713-721, 1994
28. Pattinson RC, Norman K, Odendaal HJ: The role of Doppler velocim-
etry in the management of high risk pregnancies. Br J Obstet Gynaecol
101:114-120, 1994
29. Guzman ER, Vintzileos AM, Martins M, et al: The efficacy of individual
computer heart rate indices in detecting acidemia at birth in growth-
restricted fetuses. Obstet Gynecol 87:969-974, 1996
30. Serra V, Moulden M, Bellver J, et al: The value of the short-term fetal
heart rate variation for timing the delivery of growth-retarded fetuses.
Br J Obstet Gynaecol 115:1101-1107, 2008
31. Amorim MM, Santos LC, Faundes A: Corticosteroid therapy for pre-
vention of respiratory distress syndrome in severe preeclampsia. Am J
Obstet Gynecol 180:1283-1288, 1999
32. Schiff E, Friedman SA, Kao L, et al: The importance of urinary excretion
during conservative management of severe preeclampsia. Am J Obstet
Gynecol 175:1313-1316, 1996
33. Hall DR, Odendaal HJ, Stein DW, et al: Urinary protein excretion and
expectant management of early onset, severe preeclampsia. Int J Gy-
necol Obstet 77:1-6, 2002
34. Chua S, Redman CWG: Prognosis for preeclampsia complicated by 5g
or more of proteinuria in 24 hours. Eur J Obstet Gynecol Reprod Biol
43:9-12, 1992
35. Newman MG, Robichaux AG, Stedman CM, et al: Perinatal outcomes in
preeclampsia that is complicated by massive proteinuria. Am J Obstet
Gynecol 188:264-268, 2003
36. Ganzevoort W, Rep A, Bonsel GJ, et al: A randomised controlled trial
comparing two temporising management strategies, one with and one
without plasma volume expansion, for severe and early onset pre-
eclampsia. Br J Obstet Gynaecol 112:1358-1368, 2005
37. Chari RS, Friedman SA, O’Brien JM, et al: Daily antenatal testing in
women with severe preeclampsia. Am J Obstet Gynecol 17:1207-1210,
1995
38. Hall DR, Odendaal JH, Steyn DW: Delivery of patients with early onset,
severe preeclampsia. Int J Gynecol Obstet 74:143-150, 2001
39. Alanis MC, Robinson CJ, Hulsey TC, et al: Early onset severe pre-
eclampsia: Induction of labor vs. elective Cesarean delivery and neona-
tal outcomes. Am J Obstet Gynecol 199:262.e1-262.e6, 2008