Original Research
High-Dose Methylprednisolone to Prevent
Platelet Decline in Preeclampsia
A Randomized Controlled Trial
Olivier Pourrat, MD, PhD, Marie Dorey, MD, Stphanie Ragot, PharmD, PhD, Astrid de Hauteclocque,
Philippe Deruelle, MD, PhD, Michel Dreyfus, MD, PhD, and Fabrice Pierre, MD, PhD
OBJECTIVE: To evaluate whether early administration of
high-dose methylprednisolone limits the fall of platelets
in preeclampsia.
METHODS: A randomized trial of 180 mg methylpred-
nisolone or placebo administered in divided doses over
36 hours was conducted in women admitted for pre-
eclampsia and platelet counts below 1503109/L in four
French academic centers. Patients were not included
when platelet counts were below 503109/L or when
immediate delivery was required. The primary study out-
come was the proportion of patients with platelet counts
above 1003109/L 36 hours after the first dose of study
medication. The total sample size needed to detect
a 23% difference in the rate of this outcome between
groups with a one-tailed a of 0.05 and 90% power was
94 patients.
RESULTS: Thirty-six patients were randomly assigned to
receive methylprednisolone and 34 placebo between
October 2007 and May 2011. Platelet counts above
From CHU de Poitiers, ICU and Internal Medicine, Service de Gyncologie,
Obsttrique et Mdecine de la Reproduction, Centre dInvestigation Clinique,
Poitiers, France; INSERM, CIC1402, Poitiers, France; Universit de Poitiers,
UFR Mdecine Pharmacie, Poitiers, France; Universit de Lille 2 Ple Recherche,
Lille, France; and Universit de Caen, Caen, France.
The funding source was the Ministry of Health, France (grant No. PHRC GO-46).
This study is registered with International Clinical Trials, number EudraCT2006-
004881-15-FR (PRETTY).
The funder of the study had no role in study design, data analysis, data inter-
pretation, or writing of the report. The corresponding author had full access to all
the data in the study and had final responsibility for the decision to submit for
publication.
Corresponding author: Olivier Pourrat, MD, PhD, Service de Ranimation
Mdicale et Mdecine Interne, Poitiers, France, Centre Hospitalo-Universitaire
de Poitiers, Universit de Poitiers, 2 rue de la Miltrie, BP 577, 86021 Poitiers
Cedex, France; e-mail: olivier.pourrat@chu-poitiers.fr.
Financial Disclosure
The authors did not report any potential conflicts of interest.
2016 by The American College of Obstetricians and Gynecologists. Published
by Wolters Kluwer Health, Inc. All rights reserved.
ISSN: 0029-7844/16
VOL. 128, NO. 1, JULY 2016
Copyright by The American College of Obstetricians
and Gynecologists. Published by Wolters Kluwer Health, Inc.
Unauthorized reproduction of this article is prohibited.
MD,
1003109/L at 36 hours after the first dose of study med-
ication were recorded in 30 (83%) in the active group and
29 (85%) in the placebo group (relative risk 0.98, 95%
confidence interval 0.801.20; P5.82). The only adverse
potentially study-related event was hyperglycemia in one
woman allocated to methylprednisolone.
CONCLUSION: In women with preeclampsia and plate-
let counts under 1503109/L, methylprednisolone was not
effective in maintaining platelet counts above 1003109/L.
CLINICAL TRIAL REGISTRATION: EU Clinical Trials Reg-
ister, http://clinicaltrialsregister.eu, EudraCT 2006-004881-
15-FR.
(Obstet Gynecol 2016;128:1538)
DOI: 10.1097/AOG.0000000000001470
P
reeclampsia is characterized by recent-onset arterial
hypertension together with proteinuria.1 Thrombo-
cytopenia is frequent in preeclampsia because low
platelet counts are found in half of preeclampsia2 cases
and is one of the features of hemolysis, elevated liver
enzymes, and low platelet count (HELLP) syndrome in
addition to hemolysis and elevation of liver enzymes.3
It has been argued for more than 20 years that treat-
ment with glucocorticoids could help to stabilize
HELLP syndrome before delivery and accelerate its
recovery in the postpartum period.410 However, no
randomized clinical trial (RCT) has yet demonstrated
their efficacy in this condition, as concluded by the last
Cochrane review, which, however, went on to conclude
that a consistently greater improvement in platelet
count with glucocorticoids was reason enough for fur-
ther research in this area.11
Therefore, in this trial (the PREeclampsia Throm-
bocytopenia TherapY trial), we aimed to establish
whether early and short-duration administration of
high-dose methylprednisolone would limit the fall of
platelets in women with preeclampsia and, as a result,
increase the probability that they would receive
OBSTETRICS & GYNECOLOGY 153
regional analgesia. We also assessed the occurrence of
HELLP syndrome during the evolution of the disease
as well as tolerance to methylprednisolone.
MATERIALS AND METHODS
The PREeclampsia Thrombocytopenia TherapY
study was a randomized, double-blind, placebo-con-
trolled trial conducted in four academic centers in
France (clinicaltrialsregister.eu Identifier: EudraCT
2006-004881-15-FR). The protocol was approved by
the regional ethics committee of the Region Poitou
Charentes (CPP Ouest3), France, and the study was
conducted in compliance with Good Clinical Practice
guidelines. All participants gave written informed
consent before inclusion in the trial.
All pregnant women aged 18 years or older
admitted in four academic centers at least 24 weeks
of gestation were eligible if preeclampsia, defined
according to standard criteria,1 was associated with
a platelet count less than 1503109/L but not less than
503109/L. Exclusionary criteria included pre-existing
or gestational diabetes; reported allergy to glucocorti-
coids; serum potassium less than 3.5 mmol/L;
treatment with macrolide, quinidine, or long-term
steroid regimen; possible sepsis; or severe complica-
tions requiring immediate delivery. Betamethasone
treatment, ongoing or earlier in the pregnancy (12
mg intramuscularly, two doses 24 hours apart),
given because of prematurity was not an exclusion
criterion.
Treatment assignments were determined by ran-
domized permuted blocks of size four stratified by
center. Their codes were kept in opaque, sealed, and
sequentially numbered envelopes. Patients and health
care providers (investigators, nurses, midwives) were
masked to group assignment. Data input was per-
formed by researchers at each center who were
blinded to the treatment allocation.
The syringes for the treatment were prepared by
independent nurses not involved in the patients care
and were indistinguishable in appearance and color.
The study group was given methylprednisolone
diluted in 0.9% saline and delivered with a syringe
pump over 30 minutes every 12 hours: 60 mg at the
time of randomization and 60, 40, and 20 mg 12, 24,
and 36 hours after the first dose, respectively. The
last dose was scheduled to avoid a rebound phe-
nomenon as described by OBrien.12 The total dose
of methylprednisolone was 180 mg within 36 hours.
The treatment delivered to the placebo group was
identical in terms of the number and the schedule
of injections, the procedure of administration, and
154 Pourrat et al Early Thrombopenia in Preeclampsia
Copyright by The American College of Obstetricians
and Gynecologists. Published by Wolters Kluwer Health, Inc.
Unauthorized reproduction of this article is prohibited.
the total injected volume (20 mL of 0.9% saline at
each injection).
All four injections had to be administered even if
an emergency cesarean delivery was performed
during the first 36 hours of the study. The trial was
interrupted for women developing an unfavorable
event such as a severe complication of preeclampsia,
sepsis, hyperglycemia greater than 250 mg/dL, hypo-
kalemia less than 2.5 mmol/L, or a platelet count
below the level of 503109/L needing platelet trans-
fusion. The primary outcome was the proportion of
patients whose platelet counts were above 1003109/L
36 hours after the first injection of the study drug. We
chose this threshold because regional anesthesia is
usually considered as safe if platelet counts are above
1003109/L.9
The secondary outcomes were platelet counts and
serum aspartate aminotransferase (AST), alanine
transaminase (ALT), and lactic dehydrogenase
(LDH) levels at predefined times (6, 12, 18, 24, and
36 hours after the first dose of study medication) and
their absolute variations; the incidence of HELLP
syndrome during the first 36 hours after randomiza-
tion; the rates of regional analgesia and of platelet
transfusion; maternal morbidity; and tolerance to
treatment. HELLP syndrome was defined according
to Martins criteria as class 3 if the nadir platelet count
was between 101 and 1503109/L and liver enzymes
were elevated (AST, ALT, or both 40 international
units/L or greater) and LDH 600 international units/
L or greater; class 2 if the nadir platelet count was
between 51 and 1003109/L and liver enzymes were
elevated (AST, ALT, or both 70 international units/L
or greater) and LDH 600 international units/L or
greater; and class 1 if the platelet count was below
503109/L and liver enzymes were elevated (AST,
ALT, or both 70 international units/L or greater)
and LDH 600 international units/L or greater.13
Maternal morbidity was defined according to a com-
posite variable that included any of the following
events: the need for platelet transfusion, acute renal
failure if two successive serum creatinine level results
were greater than 1.13 mg/dL, acute liver failure
defined as a factor V less than 65%, eclampsia, dis-
seminated intravascular coagulation defined by
a serum fibrinogen level less than 0.25 g/dL, pulmo-
nary edema, or acute respiratory failure defined by an
oxyhemoglobin saturation less than 94%.
Martin et al14 reported that 62% of patients with
class 3 HELLP syndrome not receiving glucocorticoids
maintained their platelet count above 1003109/L; we
assumed that this proportion could reach 85% 36 hours
after the first dose of methylprednisolone. The total
OBSTETRICS & GYNECOLOGY
Fig. 1. Diagram showing patient flow through the trial.
Pourrat. Early Thrombopenia in Preeclampsia. Obstet Gynecol 2016.

sample size needed to detect this difference between The difference between groups in the rate of the
groups with a one-tailed a of 0.05 and 90% power primary study outcome was assessed using the x2 test. Of
was 94 patients. note, for patients transfused with platelets, the primary

Table 1. Baseline Characteristics

Characteristic Methylprednisolone (n536) Placebo (n534)

Age (y) 29.064.7 29.766.0

Gestational age (wk) 30.763.8 32.263.9
Nulliparous 24 (66) 18 (53)
Uniparous 6 (17) 7 (21)
Multiparous 6 (17) 9 (26)
Medically assisted procreation 3 (8) 5 (15)
Twins 1 (3) 2 (6)
Betamethasone for fetal lung maturity 28 (78) 23 (68)
Systolic blood pressure (mm Hg) 148613 148611
Diastolic blood pressure (mm Hg) 92615 92610
Platelet count (3109/L) 123624 124624
Platelet count 1003109/L or less at enrollment 9 (25) 8 (23)
AST (international units/L) 786109 78676
ALT (international units/L) 66678 76663
LDH (international units/L) 5946313 6166382
Total bilirubin (mg/dL) 0.5360.50 0.7260.84
Serum creatinine (mg/dL) 0.760.1 0.860.2
HELLP syndrome at enrollment 10 (28) 7 (21)
Class 2 6 (17) 4 (12)
Class 3 4 (11) 3 (9)
AST, aspartate aminotransferases; ALT, alanine aminotransferases; LDH, lactate dehydrogenase; HELLP, hemolysis, elevated liver enzymes,
and low platelet count.
Data are mean6standard deviation or n (%).
Class 2 and class 3 HELLP syndrome cases were defined according to Martins criteria.13
P..05 for all between-group differences.

VOL. 128, NO. 1, JULY 2016 Pourrat et al Early Thrombopenia in Preeclampsia 155

Copyright by The American College of Obstetricians

and Gynecologists. Published by Wolters Kluwer Health, Inc.
Unauthorized reproduction of this article is prohibited.
outcome was assessed according to the last platelet count
before transfusion. Logistic regression or multivariate lin-
ear regression was used to identify potential confounders
for the relation between treatment and outcome varia-
bles. For secondary aims, analysis was performed using
Student t test (or the Mann-Whitney U test when needed)
for quantitative variables and x2 test for qualitative vari-
ables. For platelet counts, serum AST, ALT, and LDH
levels, the absolute variations between the first dose of
study medication, and the different predefined times were
calculated and compared between the two treatment
groups using a Mann-Whitney U test. Analyses were
performed by intention to treat. Statistical analyses were
performed with a one-sided significance level of 5%
using SAS 9.2.
The funder of the study had no role in study
design, data analysis, data interpretation, or writing of
the report.
RESULTS
Between October 2007 and May 2011, 72 patients
were randomly assigned to receive active methylpred-
nisolone or placebo methylprednisolone. Seventy of
them received at least one dose of the allocated
treatment and were considered in the present analysis
(Fig. 1). In the methylprednisolone group, 11 women
out of 36 (31%) did not receive all the injections
before delivery, whereas in the placebo group, 14
women out of 34 (41%) did not receive all the injec-
tions before delivery (P5.35).
Baseline characteristics were similar between the
two groups (Table 1). No significant difference was
found between the two groups in the rate of the
primary outcome: the proportion of patients with plate-
let counts above 1003109/L 36 hours after the first
dose of study medication was 83% (n530) in the

Table 2. Study Endpoints 36 Hours After the First Dose of Study Medication

Methylprednisolone (n536) Placebo (n534) P

Primary outcome
Patients with platelet count greater than 1003109/L 30 (83) 29 (85) .82
Secondary outcomes
HELLP syndrome 8 (22) 17 (50) .01
Platelet count (3109/L) 141632 132639 .27
AST (international units/L) 44650 47627 .80
ALT (international units/L) 57665 56643 .96
LDH (international units/L) 4836147 5836370 .16
Regional anesthesia 28 (78) 29 (85) .42
Maternal morbidity 6 (17) 7 (21) .67
Platelet transfusion 0 1 (3) .46
Acute renal failure 1 (3) 3 (9) .35
Eclampsia 1 (3) 1 (3) ..99
Disseminated intravascular coagulation 3 (8) 2 (9) ..99
Pulmonary edema 1 (3) 0 ..99
Acute liver failure 0 0
Acute respiratory failure 0 0
Other outcomes
Total bilirubin (mg/dL) 0.2960.24 0.4360.27 .01
Serum creatinine (mg/dL) 0.760.2 0.860.2 .30
Tolerance parameters
Glycemia (mg/dL) 113629 103620 .13
Kalemia (mmol/L) 4.360.3 4.360.4 .99
Difficulty of wound healing 2 (6) 0 (0) .49
Obstetric and neonatal parameters
Cesarean delivery 28 (78) 22 (65) .23
Gestational age at birth (weeks of gestation) 31.3 (3.7) 32.7 (3.7) .13
5-min Apgar score510 17 (55) 20 (64) .44
Birth weight (g) 1,5146819 1,8136852 .15
Platelet count at cord (3109/L) 183673 205658 .33
HELLP, hemolysis, elevated liver enzymes, and low platelet count; AST, aspartate aminotransferase; ALT, alanine aminotransferase; LDH,
lactate dehydrogenase.
Data are n (%) or mean6standard deviation.
Regional anesthesia included either epidural anesthesia or rachianesthesia.
Maternal morbidity was defined according to a composite variable including the need for platelet transfusion, acute renal failure, eclampsia,
disseminated intravascular coagulation, pulmonary edema, acute liver failure, and acute respiratory failure.
Class 2 and class 3 HELLP syndrome cases were defined according to Martins criteria.13

156 Pourrat et al Early Thrombopenia in Preeclampsia OBSTETRICS & GYNECOLOGY

Copyright by The American College of Obstetricians

and Gynecologists. Published by Wolters Kluwer Health, Inc.
Unauthorized reproduction of this article is prohibited.
methylprednisolone group and 85% (n529) in the pla-
cebo group (relative risk 0.98, 95% confidence interval
0.801.20; P5.82) (Table 2). This difference was still not
significant after adjustment for pertinent covariables.
Platelet counts, serum ALT, and LDH levels did
not differ between methylprednisolone and placebo
groups at any predefined time (Fig. 2). Regarding
serum AST, levels were significantly higher in the
placebo group 12, 18, and 24 hours after the first
dose of study medication. Taking into account the
variation of these four parameters from baseline to
other predefined times, women receiving methyl-
prednisolone had a significantly greater increase of
platelet counts than women receiving placebo both
within the first 6 hours (P5.03) and within the first 24
hours (P5.04) (Appendix 1, available online at
http://links.lww.com/AOG/A811). These differen-
ces persisted after adjustment on treatment for fetal
lung maturity and on figures of platelet counts at
baseline (P5.01 for variations within the first 6 hours
and P5.02 for variations within the first 24 hours).
Among the 53 women who had no HELLP
syndrome at baseline, none among the 26 of the
methylprednisolone group developed HELLP syn-
drome during the study period, whereas 11 of 27 of
the placebo group did (4, 3, 1, 2, and 1 cases 6, 12,
18, 24, and 36 hours after the first dose of study
medication, respectively; P,.001). However, con-
sidering each criterion defining HELLP syndrome
in this subgroup, LDH level above 600 international
units/L was the only parameter to be statistically
different between the two arms (6 [23%] among the
26 women receiving methylprednisolone and 19
[70%] among the 27 women receiving placebo;
P5.007).
The percentage of women receiving regional
anesthesia did not differ significantly between the
treatment groups nor did maternal morbidity
(Table 2). The need for platelet transfusion did not
differ significantly between groups. The only
adverse potentially study-related event was
hyperglycemia in one woman allocated to
methylprednisolone.
Regarding the neonates, gestational age was
similar as well as birth weight, Apgar score of
neonates, and umbilical artery and vein pH values.

Fig. 2. Trajectories of (A) platelet counts, (B) lactate dehydrogenase (LDH), (C) serum aspartate aminotransferase (AST), and
(D) alanine aminotransferase (ALT) levels over time according to treatment group. Aspartate aminotransferase values were
significantly lower in the methylprednisolone group than in the placebo group at hours 12, 18, and 24. Hour 0 indicates
time of the first injection. Doses of methylprednisolone were 60 mg at hours 0 and 12, 40 mg at hour 24, and 20 mg at hour
36. Data are means and standard deviations integrating all available values at the prespecified times from hour 0 to hour 36.
Red solid lines indicate the methylprednisolone group; blue dotted lines indicate the placebo group. Comparison between
methylprednisolone and placebo: *P,.05, P,.01.
Pourrat. Early Thrombopenia in Preeclampsia. Obstet Gynecol 2016.

VOL. 128, NO. 1, JULY 2016 Pourrat et al Early Thrombopenia in Preeclampsia 157

Copyright by The American College of Obstetricians

and Gynecologists. Published by Wolters Kluwer Health, Inc.
Unauthorized reproduction of this article is prohibited.
DISCUSSION
In this double-blind, placebo-controlled randomized
clinical trial we found that early intravenous adminis-
tration of high-dose methylprednisolone had no signif-
icant effect on platelet counts 36 hours after the first
injection in women with preeclampsia and platelet
counts between 150 and 503109/L. Also, methylpred-
nisolone did not have an effect on the rate of received
regional anesthesia or the need for platelet transfusion.
Administration of high-dose of methylprednisolone
over a short period of time was well tolerated.
Of the three criteria that define HELLP syndrome
according Martin et al13 (LDH 600 international
units/L or greater and elevated liver enzymes [AST,
ALT, or both 40 international units/L or greater] and
platelet count less than 1503109/L), a high level of
LDH (600 international units/L or greater) was the
only criterion that was significantly different between
the methylprednisolone and placebo groups. Effects
on LDH have been previously reported to be linked
with an action of glucocorticoids on inflammatory
cytokines.15 A RCT is currently in progress to test
the hypothesis that methylprednisolone is beneficial
when the platelet count is below 503109/L.16
The main limitation of our study is that we did
not reach our planned sample size. A lack of funding
to continue the trial after 43 months of study and
inclusion of 72 patients precluded ongoing enrollment
to our planned sample size of 94 patients.
In conclusion, high-dose methylprednisolone had
no significant effect on platelet counts 36 hours after
the first injection in women with preeclampsia and
platelet counts between 1503109/L and 503109/L.
REFERENCES
1. American College of Obstetricians and Gynecologists. Hyper-
tension in pregnancy. Washington, DC: American College of
Obstetricians and Gynecologists; 2013. Available at: http://
www.acog.org/Resources-And-Publications/Task-Force-and-
Work-Group-Reports/Hypertension-in-Pregnancy. Retrieved
May 19, 2016.
2. McCrae KR. Thrombocytopenia in pregnancy: differential diag-
nosis, pathogenesis, and management. Blood Rev 2003;17:714.
3. Weinstein L. Syndrome of hemolysis, elevated liver enzymes,
and low platelet count: a severe consequence of hypertension in
pregnancy. Am J Obstet Gynecol 1982;142:15967.
4. Martin JN Jr. Milestones in the quest for best management of
patients with HELLP syndrome (microangiopathic hemolytic
158 Pourrat et al Early Thrombopenia in Preeclampsia
Copyright by The American College of Obstetricians
and Gynecologists. Published by Wolters Kluwer Health, Inc.
Unauthorized reproduction of this article is prohibited.
anemia, hepatic dysfunction, thrombocytopenia). Int J Gynae-
col Obstet 2013;121:2027.
5. Magann EF, Bass D, Chauhan SP, Sullivan DL, Martin RW,
Martin JN Jr. Antepartum corticosteroids: disease stabilization
in patients with the syndrome of hemolysis, elevated liver en-
zymes, and low platelets (HELLP). Am J Obstet Gynecol 1994;
171:114853.
6. Magann EF, Perry KG Jr, Meydrech EF, Harris RL,
Chauhan SP, Martin JN Jr. Postpartum corticosteroids: acceler-
ated recovery from the syndrome of hemolysis, elevated liver
enzymes, and low platelets (HELLP). Am J Obstet Gynecol
1994;171:11548.
7. Martin JN Jr, Perry KG Jr, Blake PG, May WA, Moore A,
Robinette L. Better maternal outcomes are achieved with dexa-
methasone therapy for postpartum HELLP (hemolysis, elevated
liver enzymes, and thrombocytopenia) syndrome. Am J Obstet
Gynecol 1997;177:10117.
8. Martin JN Jr, Thigpen BD, Rose CH, Cushman J, Moore A,
May WL. Maternal benefit of high-dose intravenous corticoste-
roid therapy for HELLP syndrome. Am J Obstet Gynecol 2003;
189:8304.
9. Martin JN Jr, Rose CH, Briery CM. Understanding and man-
aging HELLP syndrome: the integral role of aggressive gluco-
corticoids for mother and child. Am J Obstet Gynecol 2006;
195:91434.
10. van Runnard Heimel PJ, Huisjes AJ, Franx A, Koopman C,
Bots ML, Bruinse HW. A randomised placebo-controlled trial
of prolonged prednisolone administration to patients with
HELLP syndrome remote from term. Eur J Obstet Gynecol
Reprod Biol 2006;128:18793.
11. Woudstra DM, Chandra S, Hofmeyr GJ, Dowswell T. Cortico-
steroids for HELLP (hemolysis, elevated liver enzymes, low
platelets) syndrome in pregnancy. The Cochrane Database of
Systematic Reviews 2010, Issue 9. Art. No.: CD008148. DOI:
10.1002/14651858.CD008148.pub2.
12. OBrien JM, Milligan DA, Barton JR. Impact of high-dose cor-
ticosteroid therapy for patients with HELLP (hemolysis, ele-
vated liver enzymes, and low platelet count) syndrome. Am J
Obstet Gynecol 2000;183:9214.
13. Martin JN Jr, Rinehart BK, May WL, Magann EF, Terrone DA,
Blake PG. The spectrum of severe preeclampsia: comparative
analysis by HELLP (hemolysis, elevated liver enzyme levels,
and low platelet count) syndrome classification. Am J Obstet
Gynecol 1999;180:137384.
14. Martin JN Jr, Blake PG, Perry KG Jr, McCaul JF, Hess LW,
Martin RW. The natural history of HELLP syndrome: patterns
of disease progression and regression. Am J Obstet Gynecol
1991;164:15009.
15. Wallace K, Martin JN Jr, Tam Tam K, Wallukat G, Dechend R,
Lamarca B, et al. Seeking the mechanism(s) of action for corti-
costeroids in HELLP syndrome: SMASH study. Am J Obstet
Gynecol 2013;208:380.e18.
16. Katz L, Amorim M, Souza JP, Haddad SM, Cecatti JG, CO-
HELLP Study Group. COHELLP: collaborative randomized
controlled trial on corticosteroids in HELLP syndrome. Reprod
Health 2013;10:28.
OBSTETRICS & GYNECOLOGY