wallace 2013

dsticpl

sp

Research www.AJOG.org

OBSTETRICS

Seeking the mechanism(s) of action for corticosteroidsin HELLP syndrome: SMASH studyKedra Wallace, PhD; James N. Martin Jr, MD; Kiran Tam Tam, MD; Gerd Wallukat, PhD;Ralf Dechend, PhD; Babbette Lamarca, PhD; Michelle Y. Owens, MD

INTRODUCTION: Administration of dexamethasone to the hemolysis,elevated liver enzymes, and low platelet count (HELLP) syndrome pa-tients (10 mg intravenously [IV] every 12 hours) shortens the diseasecourse and reduces maternal morbidity in patients treated at the Univer-sity of Mississippi Medical Center (UMMC), associated with this severeform of preeclampsia. However, the pathophysiological mechanismsinvolved with this intervention remain unclear.

OBJECTIVE: We sought to investigate the potential role of IV dexametha-sone to restore the imbalance among antiangiogenic and inflammatory fac-tors known to be significantly elevated in women with HELLP syndrome.

STUDY DESIGN: This was a single-center prospective study of womeniagnosed with HELLP syndrome who were treated for IV dexametha-one at UMMC. Blood was drawn prior to dexamethasone administra-ion and again 12 and 24 hours after the initial dexamethasone admin-stration. Enzyme-linked immune assays were used to measureirculating inflammatory cytokines and antiangiogenic factors. A re-eated-measures analysis of variance was used to analyze the data col-

ected before, after, and during dexamethasone administration.

Obstet Gynecol 2013;208:380.e1-8.

dgfnwH

0002-9378/$36.00 • © 2013 Mosby, Inc. All rights reserved. • http://dx.doi.org/10.10

380.e1 American Journal of Obstetrics & Gynecology MAY 2013

RESULTS: Seventeen women with HELLP syndrome were enrolled inthis study. Dexamethasone significantly decreased evidence of hemoly-sis (P � .002) and liver enzymes (P � .003), and significantly in-creased platelets (P � .0001) within 24 hours of administration. Circu-lating interleukin-6 levels after 24 hours were decreased (P � .001);soluble fms-like tyrosine kinase-1 and soluble endoglin were also sig-nificantly decreased by 24 hours after dexamethasone administration(P � .002 and P � .004, respectively). There were no significant differ-ences in circulating levels of placental growth factor (P � .886) due todexamethasone administration. Angiotensin II receptor autoantibodylevels were unchanged by dexamethasone administration.

CONCLUSION: We conclude that 1 important mechanism of dexamethasoneadministration is to blunt the release of both antiangiogenic and inflammatoryfactors suggested to play role in the pathophysiology of HELLP syndrome.

Key words: HELLP (hemolysis, elevated liver enzymes, and lowplatelet count) syndrome, inflammatory cytokines, severe
preeclampsia, soluble endoglin, soluble fms-like tyrosine kinase-1
Cite this article as: Wallace K, Martin JN Jr, Tam Tam K, et al. Seeking the mechanism(s) of action for corticosteroids in HELLP syndrome: SMASH study. Am J

Hemolysis, elevated liver enzymes,and low platelet (HELLP) syn-

drome, a form of severe preeclampsia/eclampsia, affects 10% of women withpreeclampsia in the United States andserves as a major contributor to mater-nal-perinatal morbidity and mortali-ty.1-6 Recent evidence has demon-trated an imbalance betweenroangiogenic factors (ie, vascular en-

From the Division of Maternal-Fetal Medicine, DeUniversity of Mississippi Medical Center, Jacksoand Owens), and HELIOS Clinic, Experimental aCenter, Berlin, Germany (Drs Wallukat and Dech

Received Dec. 11, 2012; revised Jan. 7, 2013; a

This study was supported by the Division of Matand Gynecology, University of Mississippi Medic

The authors report no conflict of interest.

Presented as a poster at the 76th Annual Confeand Gynecologists, Wailea, Maui, HI, Oct. 23-25

Reprints not available from the authors.

othelial growth factor and placentalrowth factors) and antiangiogenicactors (ie, soluble fms-like tyrosine ki-ase [sFlt-1] and soluble endoglin) inomen with preeclampsia and/orELLP syndrome.7-9

During human preeclampsia and in an-imal models of the disease, sFlt-1 and sol-ubleendoglinareassociatedwith increasedimmune activation and elevations in in-

rtment of Obstetrics and Gynecology,S (Drs Wallace, Martin, Tam Tam, Lamarca,

Clinical Research Center, Max-Delbrueck).

pted Jan. 29, 2013.

al-Fetal Medicine, Department of Obstetricsenter.

e of the Central Association of Obstetricians09.

16/j.ajog.2013.01.049

flammatory cytokines tumor necrosis fac-tor alpha (TNF-�) and interleukin-6 (IL-6).10-12 Furthermore, many studies haveshown a close association between agonis-tic autoantibodies to the angiotensin II re-ceptor (AT1-AA) and sFlt-1.13-17

The Xia laboratory recently demon-strated that the severity of preeclampsiawas tightly associated with the increase inAT1-AA and sFlt-1, thus indicating thatthese factors could potentially serve as bio-markers to predict the severity of pre-eclampsia during pregnancy.18 However,studies demonstrating high levels ofAT1-AA in women with HELLP syn-drome are lacking. Furthermore, studieslinking antiangiogenic factors and solubleinflammatory markers during HELLPsyndrome are sparse.

In recent years, many studies have re-ported conflicting results for the benefi-cial role of corticosteroids in the setting

pan, Mndend

cce

ernal C

renc, 20

of HELLP syndrome.6,15,19-21 The use of

http://dx.doi.org/10.1016/j.ajog.2013.01.049

pwtsaw

dmtdodflhp

(laIc

t

Idmd

wntftc

www.AJOG.org Obstetrics Research

dexamethasone to improve maternaloutcome has been used in our tertiarycare hospital for 2 decades. Recently wereported a study of 190 patients withHELLP syndrome who were treated withwhat is termed the Mississippi Protocol,an approach to patient management thatincludes intravenous (IV) dexametha-sone administered immediately upon di-agnosis of HELLP syndrome, and dem-onstrated a halt to disease progression aswell as a reduction in new major mater-nal morbidity.5,6,20 Indeed, given the

resence of inflammation in womenith HELLP syndrome, we hypothesized

hat IV administration of dexametha-one could decrease the inflammationnd/or antiangiogenic factors associatedith HELLP syndrome.Therefore, the aim of this study was to

emonstrate that intravenous dexa-ethasone, as the core component of

herapy for patients with HELLP syn-rome, improves the clinical parametersf affected patients in association withecreased antiangiogenic factors and in-ammatory cytokines. Both of theseave been suggested to play a role in theathophysiology of HELLP syndrome.

MATERIALS AND METHODSStudy populationThis was a prospective trial that involvedpregnant patients with the diagnosis ofHELLP syndrome admitted to the Uni-versity of Mississippi Medical Center(Jackson, MS) between 2007 and 2011.

Inclusion criteria were patients pre-senting in the antepartum, intrapartum,or postpartum period with a primary in-dication of a HELLP syndrome. Patientsare described as having met complete orincomplete HELLP syndrome criteriausing the Mississippi classification sys-tems for classes 1, 2, and 3 of HELLPsyndrome.22 Class 1 HELLP syndromeplatelet count �50,000/�L, total serumactic dehydrogenase [LDH] �600 IU/L,spartate aminotransferase [AST] �70U/L), class 2 HELLP syndrome (plateletount 50,000 of �100,000/�L, total se-

rum LDH �600 IU/L, AST �70 IU/L),and class 3 HELLP is characterized by(platelet count 100,000 of �150,000/�L,
otal serum LDH �600 IU/L, AST �70
U/L). Partial or incomplete HELLP wasefined as the presence of only 2 of the 3ajor laboratory criteria to establish a

iagnosis of complete HELLP syndrome.Exclusion criteria included treatmentith corticosteroids for any other mater-al/fetal indication, having received cor-

icosteroids within 10 days of diagnosisor any reason or having an active sys-emic infection that would contraindi-ate the use of corticosteroids.

Management of HELLP syndromeHELLP syndrome was managed follow-ing the Mississippi Protocol which in-cludes three major components: (1)magnesium sulfate by intravenous infu-sion primarily for eclampsia seizure pro-phylaxis and reduction of systemic vas-cular resistance through delivery and upto 24 hours postpartum with durationdependent on disease acuity and whet-her HELLP syndrome is first-evidencedpostpartum; (2) blood pressure controlusing oral or intravenous medication(hydralazine or labetol) to sustain sys-tolic blood pressure less than 160 mm Hgand diastolic blood pressure less than100 mm Hg; and (3) IV dexamethasone10 mg at 12 hour intervals until pla-telet normalization is trending toward100,000 �L at which time 5 mg dexa-methasone at 12 hour intervals is admin-istered twice intravenously before cessa-tion to minimize the risk of reboundthrombocytopenia.20 Importantly, theintravenous dexamethasone is initiatedas soon as the diagnosis of HELLP syn-drome is made in patients with class 1 or2 HELLP syndrome or in patients withclass 3 or partial HELLP syndrome com-plicated by eclampsia, severe systolic hy-pertension, severe epigastric pain, or arapidly deteriorating clinical status.5,6,20

Laboratory methodsData collection included maternal clini-cal parameters (symptoms of headaches,right upper abdominal pain, blood pres-sure), measures of maternal morbidity(cardiopulmonary, hematocoagulation,hepatorenal, infectious, obstetric), labo-ratory values (platelet count, hematocrit,LDH, creatinine, AST, uric acid, cyto-kine, and vascular factors), and fetal
outcome variables (gestational age at de-
MAY 2013 Americ

livery, birthweight, Apgar scores, re-quirement for mechanical ventilation,other measures of neonatal morbidity).For the purposes of the current study,blood was drawn at the time of study ad-mission (prior to corticosteroid admin-istration) and at 12 and 24 hours afterinitial dexamethasone administrationbefore the next dose was given.

Determination of cytokineand AT1-AA productionPlasma was collected and measured forangiogenic factors: sFlt-1, soluble en-doglin, placental growth factor (PlGF)and inflammatory cytokines: IL-6,TNF-�. Commercial ELISA kits avail-able from R&D Systems (Quantikine;R&D Systems, Minneapolis, MN) wereused according to the manufactu-rer’s protocol. Sera immunoglobulinfractions were isolated from HELLPpatients as previously described.23 AT1-AA activity was measured utilizing a bio-assay that evaluates the beats/minute(bpm) of neonatal cardiomyocytes inculture.23,24

Statistical analysisData are expressed as mean � SEM. Arepeated-measures analysis of variancewas used to determine differences inclinical and circulating parameters be-fore, during, and after dexamethasoneadministration, with a paired Student ttest being used for post hoc analysis. AStudent’s t test was also used to deter-mine differences in clinical laboratoryparameters between women with ante-partum vs postpartum HELLP. Spear-man calculation was used to determinewhether there was a correlation betweenthe final platelet count and the numberof dexamethasone administrations be-fore delivery. Percent change was calcu-lated as ([value post dexamethasone-Time before dexamethasone]/valuebefore dexamethasone) � 100. Valueswere considered statistically significantat P � .05.

RESULTSPatient population and demographicsDuring the study period 2007-2011, 19 fe-
male patients with a diagnosis of HELLP
an Journal of Obstetrics & Gynecology 380.e2

(mra1Fmep(nc

a

db1tphdmd

Research Obstetrics www.AJOG.org

syndrome meeting eligibility criteria con-sented for enrollment in the current study.Of these, 17 completed the study, withblood drawn at baseline and approxi-mately 12 and 24 hours after the initial IVdexamethasone administration.

Women enrolled in this study weresimilar regarding baseline demographicand pregnancy characteristics, with theexception of maternal race (Table 1).Sixteen of 17 (94.1%) of the study par-ticipants were self-reported AfricanAmerican. Approximately 71% of the pa-tients had cesarean deliveries in which themost common indications were electiverepeat, failure to progress, and nonreassur-ing fetal heart rates. There was 1 patientwith a previous history of eclampsia, whoalso had eclampsia in the current study.Seven women (41%) received both dosesof dexamethasone and had blood drawnprior to delivery, 2 women (12%) receivedboth doses of dexamethasone and hadtheir last blood draw within 3 hours of de-livery, 3 women (18%) received 1 dose ofdexamethasone before delivery and hadher last dose and last 2 blood draws afterdelivery. Five patients (29%) enrolled inthe study developed HELLP syndromeduring the postpartum period and did notreceive any dexamethasone prior todelivery.

Biochemical parametersof HELLP syndromeTo determine whether there were differ-ences in the clinical laboratory values ofHELLP syndrome between women whodeveloped HELLP in the antepartum vspostpartum period, we compared LDH,AST, and platelet values at study enroll-ment. At the time of study enrollment,there were no significant differences inplatelet counts (P � .114) or LDH (P �.451) or AST levels (P � .182, Figure 1).Because there were no significant differ-ences in laboratory values betweengroups, all patients were analyzed to-gether, independent of when they devel-oped HELLP syndrome.

Dexamethasone improves clinicalsigns and laboratory valuesClinical findings indicated that systolic anddiastolic blood pressures were significantly
decreased at 12 (P� .020; P� .011) and 24 w
380.e3 American Journal of Obstetrics & Gynecolo

P � .002; P � .014) hours after IV dexa-ethasone administration (Table 2). With

egard to laboratory values, IV dexameth-sone significantly increased platelets after2 (P � .001) and 24 hours (P � .0001;igure 2, A). The number of times dexa-ethasone was administered before deliv-

ry did not have a significant effect on thelatelet count at the 24 hour blood drawP � .476; data not shown). There wasot a significant difference in plateletount at 24 hours between the women

TABLE 1Maternal demographic data

Demographic

n...................................................................................................................

Age, y..........................................................................................................

Median (range)...................................................................................................................

Race (% African American)...................................................................................................................

Primigravida...................................................................................................................

Gestational age at delivery, wks..........................................................................................................

Median (range)...................................................................................................................

Mode of delivery..........................................................................................................

Primary cesarean..........................................................................................................

Vaginal...................................................................................................................

DX chronic hypertension...................................................................................................................

Current Dx preeclampsia...................................................................................................................

Previous DX preeclampsia in prior pregnancy...................................................................................................................

Current DX of eclampsia...................................................................................................................

Previous DX of eclampsia in prior pregnancy...................................................................................................................

Class 1 HELLP syndrome...................................................................................................................



Indicators of preeclampsia/HELLP syndrome..........................................................................................................

Increasing blood pressure..........................................................................................................

Headaches unrelieved with Tylenol..........................................................................................................

Uric acid �6.0 mg/dL..........................................................................................................

AST �70 IU/L..........................................................................................................

Visual disturbance..........................................................................................................

Thrombocytopenia..........................................................................................................

Epigastric pain..........................................................................................................

Proteinuria..........................................................................................................

Disseminated intravascular coagulation...................................................................................................................

Tylenol; McNEIL-PPC, Inc., Fort Washington, PA.AST, aspartate aminotransferase; DX, diagnosis; HELLP, hem

Wallace. The SMASH study. Am J Obstet Gynecol 2013.

ho received both doses of dexameth- r

gy MAY 2013

sone before delivery (105,500 � 5500�L) and the women who received both

oses and had their final blood drawefore delivery (P � .804; 111,714 �2,222 �L). There was not a correla-ion between the percent increase inlatelet count from baseline to 24ours after the initial dexamethasoneose and the number of times dexa-ethasone was administered before

elivery (r � 0.4, P � .750).Hematocrit significantly decreased in

% (x/n)

17..................................................................................................................

24 � 1.5..................................................................................................................

23 (18–40)..................................................................................................................

94.1% (16/17)..................................................................................................................

35.2% (6/17)..................................................................................................................

31.9 � 1.2..................................................................................................................

32.6 (20.4–39)..................................................................................................................

..................................................................................................................

33.3% (4/12)..................................................................................................................

41.6% (5/12)..................................................................................................................

29.4 (5/17)..................................................................................................................

58.8% (10/17)..................................................................................................................

18.2% (2/11)..................................................................................................................

11.7% (2/17)..................................................................................................................

9.0% (1/11)..................................................................................................................

35.2% (6/17)..................................................................................................................

47.0% (8/17)..................................................................................................................

17.6% (3/17)..................................................................................................................

..................................................................................................................

70.5% (12/17)..................................................................................................................

11.7% (2/17)..................................................................................................................

23.5% (4/17)..................................................................................................................

82.3% (14/17)..................................................................................................................

0% (0/17)..................................................................................................................

17.6% (3/17)..................................................................................................................

35.2% (6/17)..................................................................................................................

47.0% (8/17)..................................................................................................................

5.8% (1/17)..................................................................................................................

, elevated liver enzymes, and low platelet count.

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

.........

olysis

esponse to dexamethasone (P � .0004)

6Ptlndlbm(maF

btesmewlsd


at 12 hours and (P � .0001) after 24hours (Figure 2, B). Serum LDH andAST levels both decreased 12 (P � .004;P � .006) and 24 hours (P � .002; P �.003) after dexamethasone administra-tion and continued to significantly de-crease after an additional dexametha-sone treatment (P � .005; P � .005;Figure 2, C and D). There was not a sig-nificant change in creatinine levels be-cause of dexamethasone (P � .611; Table2). Uric acid increased (P � .05) 24

FIGURE 1Platelet count and LDH and AST lev

The platelet count and LDH and AST levels at thebetween women who developed HELLP anteparpostpartum.AST, aspartate aminotransferase; HELLP, hemolysis, elevated live


TABLE 2Comparison of data before, during,

Parameter

Systolic blood pressure, mm Hg (range)...................................................................................................................

Diastolic blood pressure, mm Hg (range)...................................................................................................................

Platelets, �L...................................................................................................................

AST, IU/L...................................................................................................................

LDH, IU/L...................................................................................................................

Creatinine, mg/dL...................................................................................................................

Uric acid, mg/dL...................................................................................................................

Proteinuria indicated...................................................................................................................

Data are represented as the mean � SEM.AST, aspartate aminotransferase; LDH, lactic dehydrogenase.a P � .05 compared with parameter value at enrollment; b P


hours after dexamethasone administra-tion but did not significantly changeduring the first 12 hours of dexametha-sone treatment (P � .098; Table 2). Fifty-three percent of patients had proteinuriaat admission (Table 2).

Antiangiogenic factors are decreasedin response to dexamethasoneCirculating soluble endoglin was signifi-cantly decreased by 32.3% at12 hours(40.65 � 5.07 ng/mL; P � .034) and by

at the time of enrollment

e of enrollment were not statistically significantcompared with women who developed HELLP

ymes, and low platelet count; LDH, lactic dehydrogenase.

d after 24 hours of dexamethasone

rollment� 17)

12 hours after(n � 17)

156.9 � 4.7 (128–190) 138.4 � 3.4.........................................................................................................................

93.5 � 3.6 (69–115) 82.2 � 3.9.........................................................................................................................

,647 � 5576 93,706 � 5818.........................................................................................................................

195.5 � 43.24 96.29 � 16.........................................................................................................................

1737 � 152.7 1408 � 98.7a

.........................................................................................................................

0.75 � 0.05 0.75 � 0..........................................................................................................................

4.95 � 0.49 5.67 � 0..........................................................................................................................

53% (9/17) —.........................................................................................................................

5 compared with parameter value 12 hours after administration o

MAY 2013 Americ

2.8% at 24 hours (21.02 � 4.71 ng/mL;� .004) after dexamethasone adminis-

ration compared with soluble endoglinevels at study enrollment (54.80 � 6.08g/mL; Figure 3, A). Administration ofexamethasone decreased circulating

evels of the antiangiogenic factor, sFlt-1y 25% at 12 hours (3839 � 701.4 pg/L; P � .020) and by 43.6% at 24 hours

2814 � 578 pg/mL; P � .002) after ad-inistration compared with sFlt-1 levels

t study enrollment (5213 � 786 pg/mL;igure 3, B).An imbalance in sFlt-1 and PlGF has

een suggested to be useful in predictinghe clinical course of diseases such as pre-clampsia and HELLP syndrome.25 Asuch, we determined the effect of dexa-

ethasone administration on plasma lev-ls of PlGF and the sFlt-1/PlGF ratio. Thereere no significant differences in the circu-

ating levels of PlGF (P � .886) or on theFlt-1/PlGF ratio (P � .831) because ofexamethasone administration.

Dexamethasone decreasesinflammatory cytokines in womenwith HELLP syndromeCirculating IL-6 levels were significantlydecreased 24 hours after dexamethasoneadministration compared with initialIL-6 levels (50.34 � 5.08 vs 72.55 � 5.17pg/mL; P � .001; Figure 4, A). Therewere no significant differences in IL-6levels 12 hours after the initial dexa-methasone administration (P � .133).

inistration

24 hours after(n � 17)

0–155)a 134.4 � 2.1 (125–145)a..................................................................................................................

–107)a 79.4 � 3.1 (65–95)a..................................................................................................................

119,118 � 6451a,b

..................................................................................................................

63.41 � 8.9a,b

..................................................................................................................

1193 � 95.6a,b

..................................................................................................................

0.78 � 0.05..................................................................................................................

5.94 � 0.52a

..................................................................................................................

—..................................................................................................................

xamethasone.

els

timtum

r enz

an adm

En(n

(12......... .........

(60......... .........

80 a

......... .........

.1a

......... .........

......... .........

05......... .........

45......... .........

......... .........

� .0 f de


t3

b

b

tptp

pahwttetiad

dts


Dexamethasone decreased circulatingTNF� levels 24 hours after administra-ion (134.3 � 46.11 initially vs 115 �2.18 pg/mL; P � .274; Figure 4, B);

however, this was not statisticallysignificant.

Dexamethasone has no effecton AT1-AA productionThe AT1-AA has been found to be in-creased in women with preeclampsiacompared with normal pregnant andnonpregnant women. We sought to de-termine whether the AT1-AA is alsopresent in women with HELLP syn-drome and whether dexamethasone ad-ministration could decrease the produc-tion of the autoantibody. There were nosignificant differences in AT1-AA pro-duction at either 12 hours (28.5 � 1.9bpm; P � .908) or 24 hours (29.6 � 1

pm; P � .711) after dexamethasone ad-ministration compared with AT1-AAlevels at study enrollment (28.80 � 1.3

pm; Figure 4, C).

Newborn outcomesThe average birthweight was 2916 �1381 g, the average gestational age was31.89 � 1.24 weeks, with a 1 minute Ap-gar average of 6 and a 5 minute Apgaraverage of 8. There were 4 cases of respi-ratory distress syndrome (27%), with 1requiring intubation, 1 case (7%) oftransient tachypnea of the newborn, and2 cases of neonatal thrombocytopenia(13%). No newborn developed intraven-tricular hemorrhage, retinopathy of pre-maturity, or necrotizing entercolitis.There were 2 stillborn infants in thisstudy.

COMMENTThe current study further expands onthe reported experience at our institu-tion with HELLP syndrome by demon-strating that implementation of the Mis-sissippi Protocol slows the diseaseprogression, improves already presentmaternal morbidity, and lessens new-onset maternal morbidity and likely less-ens maternal mortality.5,6,20,26 Underhe Mississippi Protocol, the patient whoresents with HELLP syndrome isreated with the following: (1) antihy-
ertensives to lower blood pressure and

revent stroke and complications associ-ted with severe systolic and/or diastolicypertension; (2) magnesium sulfate,hich has been shown to reduce sys-

emic vascular resistance, provide ma-ernal neuroprotection, and preventclamptic seizure; and (3) most impor-antly, IV dexamethasone to reduce thenflammation that underlies the clinicalnd laboratory events of HELLP syn-rome.20,27-29 However, some studies

have reported that IV dexamethasoneadministration does not appear to have abeneficial effect in halting the progres-sion of HELLP syndrome and or mater-nal morbidity.19,21 However, our studyrefutes that, indicating that administra-tion of dexamethasone in addition to an-tihypertensives and magnesium sulfateimproves clinical outcomes in our pa-tient population. The hypertension asso-

FIGURE 2Platelet count and hematocrit, LDHlevels with dexamethasone admini

A, Platelet count and B, hematocrit, C, LDH, and Dexamethasone administration. Each line represehe study. Asterisk indicates P � .05 and doignificant change in parameter values between

AST, aspartate aminotransferase; LDH, lactic dehydrogenase.


ciated with HELLP syndrome decreased

gy MAY 2013

significantly upon IV dexamethasoneadministration. In addition, IV dexa-methasone produced a significant im-provement in platelet levels, hematocrit,and LDH and AST levels within 12 hoursof administration and continued to do soafter a second administration of dexa-methasone (Figure 1).

Hypertension is well reported to be as-sociated with the excessive production ofinflammatory cytokines, agonistic auto-antibodies to the angiotensin II receptor,and imbalances in antiangiogenic factorssuch as sFlt-1 and soluble endoglin inboth clinical and experimental stud-ies.11,12,25,30-32 The imbalance in anti-angiogenic factors is well established inpreeclampsia and HELLP syndrome inboth women and animal models of thesediseases.7,12,25,31,33 Venkatesha et al8

have shown that by overexpressing

nd ASTation

ST levels before, at 12 hours, and 24 hours afterlaboratory values for an individual patient across

asterisk indicates P � .005 and denotes aindicated groups.

, astr

, Antsublethe

sFlt-1 and soluble endoglin, animals

iAd

d

i


develop characteristics similar toHELLP syndrome. Although the exactmechanism by which angiogenic im-balance induces proteinuria, hyperten-sion, and inflammation is unclear, itserves as a hallmark of both HELLPsyndrome and preeclampsia.7,8,12,31,34

In the current study, IV dexametha-sone decreased both sFlt-1 and solubleendoglin, IL-6, and TNF-� after 24hours. However, there was no effect onPlGF or the ratio of sFlt-1/PlGF. Al-though administration of dexametha-sone had no effect of AT1-AA, the signif-icant effects on other soluble factorsknown to stimulate AT1-AA productionsuggest that over time administration ofdexamethasone could have a profoundeffect to suppress AT1-AA and other im-

FIGURE 3Soluble endoglin and sFlt-1response to dexamethasone

% Change in Circulating sEndoglin

12hrs 24hrs

-80

-60

-40

-20

0

*

***

% C

hang

e

% Change in Circulating sFlt-1

12hrs 24hrs

-60

-40

-20

0

*

**

% C

hang

e

A

B

Circulating antiangiogenic factors, A, soluble en-oglin and B, sFlt-1 are significantly decreased

in response to dexamethasone after 12 and 24hours. Data are represented as the percentchange between the initial soluble endoglin orsFlt-1 value and the soluble endoglin or sFlt-1value at 12 and 24 hours after dexamethasoneadministration. Asterisk indicates P � .05, dou-ble asterisk indicates P � .005, and triple aster-isk indicates P � .0005 and denotes a signifi-cant change in parameter values from theindicated time and time 0.sFlt-1, soluble fms-like tyrosine kinase.


mune factors known to play a patho-

physiological role in forms of hyperten-sion during pregnancy.

Recent studies describe the relation-ship between increased sFlt-1 and solu-ble endoglin and endothelin-1 (ET-1) inwomen and animal models of placentalischemia.31 Aggarwal et al32 suggestedthat the increase in these antiangiogenicfactors is in part responsible for the in-crease in ET-1, which is associated withpreeclampsia. Furthermore, Murphy etal31 demonstrated that sFlt-1-induced hy-pertension is mediated by activation of theET-1 pathway. Indeed, animal models ofplacental ischemia have shown that uponET-1 blockade, indicating that some of theclinical picture, because of placental isch-emia, may indeed be due to the increase inET-1.12,31,35-37 ET-1 activation is a result ofncreased inflammatory cytokines and AT1-A, which are present in both HELLP syn-rome and preeclampsia.31,32,38-41

One of the limitations in this study isthat all patients received an antihyper-tensive in addition to magnesium sul-fate. Therefore, it is hard to say whetherthe results we see are due just to dexa-methasone administration, antihyper-tensives, or the combination of all 3components used in the Mississippi Pro-tocol.20 Clinically, we see that bloodpressure control is achieved more readilyand consistently once IV dexamethasoneis initiated. It is possible that this is due tothe antiinflammatory and immuno-modulatory effects that dexamethasonehas on the endothelium and vasculatureor through a reduction in neurogenic in-flammation.42-44 Because not all of thewomen enrolled in this study were ableto complete the 24 hour dexamethasonewindow before delivery, it is importantto point out that delivery does serve as aconfound to these results.

Future clinical and animal studies willinclude matched control arms that willallow us to see the effects that antihyper-tensives and magnesium sulfate have onthe biochemical and antiangiogenicproperties of HELLP syndrome with andwithout dexamethasone. Additionalstudies are also needed to determinewhether platelet levels increase and sFltand soluble endoglin levels decrease at 24
and 48 hours after delivery in women
MAY 2013 Americ

with antepartum HELLP who have andhave not received dexamethasone.

With this study we demonstrate thatthe early administration of intravenousdexamethasone along with administra-tion of antihypertensives and magne-sium sulfate proves beneficial for treat-

FIGURE 4IL-6 and TNF-� responseto dexamethasone

Circulating inflammatory cytokines, A, IL-6 andB, TNF-�, are decreased in response to dexa-methasone after 12 and/or 24 hours. C, Produc-tion of the AT1-AA was measured and was foundto be present in women with HELLP syndrome.However, dexamethasone did not decrease pro-duction. Data are represented as mean � SEM.Asterisk indicates P � .05 and double asteriskndicates P � .005 and denotes a significantchange in parameter values between the indi-cated groups.AT1-AA, angiotensin II receptor; HELLP, hemolysis, elevated liverenzymes, and low platelet count; IL-6, interleukin-6; TNF-�,tumor necrosis factor alpha.


ing HELLP syndrome during pregnancy.



Dexamethasone improves soluble fac-tors shown to cause hypertension and in-trauterine growth restriction such asTNF-�, IL-6, sFlt-1, and soluble en-doglin. Importantly, administration ofdexamethasone within the treatmentregimen improves blood pressure, plate-lets, and liver enzyme values of womensuffering with HELLP syndrome. Al-though others dispute the usefulness ofdexamethasone during this devastatingdisease, we demonstrate that imple-menting the Mississippi Protocol fortreating HELLP patients proves benefi-cial to maternal outcome and will con-tinue to be routine in UMMC medicalpractice for improving health care forwomen and subsequently children af-fected by HELLP syndrome. f

REFERENCES1. Geary M. The HELLP syndrome. Br J ObstetGynaecol 1997;104:887-901.2. Karumanchi S, Maynard S, Stillman I, EpsteinF, Sukhatme S. Preeclampsia: a renal perspec-tive. Kidney Int 2005;67:2101-13.3. Sibai B. Diagnosis, controversies, and man-agement of the syndrome of hemolysis, ele-vated liver enzymes and low platelet count. Ob-stet Gynecol 2004;103:981-91.4. Keiser S, Owens M, Parrish M, et al. HELLPsyndrome with and without eclampsia. Am JPerinatol 2011;28:187-94.5. Martin JJ, Magann E, Isler C. HELLP syn-drome: the scope of disease and treatment. In:Thornton S, Saade G, eds. Hypertension inpregnancy. Belfort: Oxford; 2003.6. Martin JJ, Rose C, Briery C. Understandingand managing HELLP syndrome: the integral roleof aggressive glucocorticoids for mother andchild. Am J Obstet Gynecol 2006;195:914-34.7. Maynard S, Min J, Merchan J, et al. Excessplacental soluble fms-like tyrosine kinase 1(sFlt1) may contribute to endothelial dysfunc-tion, hypertension, and proteinuria in pre-eclampsia. J Clin Invest 2003;111:649-58.8. Venkatesha S, Toporsian M, Lam C, et al.Soluble endoglin contributes to the pathogene-sis of preeclampsia. Nat Med 2006;12:642-9.9. Zhou Y, McMaster M, Woo K, et al. Vascularendothelial growth factor ligands and receptorsthat regulate human cytotrophoblast survival aredysregulated in severe preeclampsia and hemo-lysis, elevated liver enzymes and low platelets syn-drome. Am J Pathol 2002;160:1405-23.10. Xu B, Thornton C, Tooher J, Hennessy A. Ex-ogenous soluble VEGF receptor-1 (sFlt-1) regu-lates Th1/Th2 cytokine production from normalplacental explants via intracellular calcium. Hyper-tens Pregnancy 2009;28:448-56.11. LaMarca B, Bennett W, Alexander B, Cock-
rell K, Granger J. Hypertension produced by

reductions in uterine perfusion in the pregnantrat: role of tumor necrosis factor-alpha.Hypertension 2005;46:1022-5.12. Parrish M, Murphy S, Rutland S, et al. Theeffect of immune factors, tumor necrosis factor-alpha, and agonistic autoantibodies to the an-giotensin II type I receptor on soluble fms-liketyrosine-1 and soluble endoglin production inresponse to hypertension during pregnancy.Am J Hypertens 2010;23:911-6.13. Visser W, Beckmann I, Bremer H, Lim H,Wallenburg H. Bioactive tumour necrosis factoralpha in pre-eclamptic patients with and withoutthe HELLP syndrome. Br J Obstet Gynaecol1994;101:1752-7.14. Tranquilli A, Landi B, Corradetti A, et al. In-flammatory cytokines patterns in the placenta ofpregnancies complicated by HELLP (hemoly-sis, elevated liver enzyme, and low platelet) syn-drome. Cytokine 2007;40:82-8.15. van Runnard Heimel P, Kavelaars A, HeijnenC, et al. HELLP syndrome is associated with anincreased inflammatory response, which may beinhibited by administration of prednisolone. Hy-pertens Pregnancy 2008;27:253-65.16. Kupferminc M, Peaceman A, Wigton T,Rehnberg K, Socol M. Tumor necrosis factor-alpha is elevated in plasma and amniotic fluid ofpatients with severe preeclampsia. Am J ObstetGynecol 1994;170:1752-7.17. Dechend R, Homuth V, Wallukat G, et al.Agonistic antibodies directed at the angiotensinII, AT1 receptor in preeclampsia. J Soc GynecolInvestig 2006;13:79-86.18. Zhou C, Ahmad S, Mi T, et al. Autoantibodyfrom women with preeclampsia induces solubleFms-like tyrosine kinase-1 production via an-giotensin type 1 receptor and calcineurin/nu-clear factor of activated T-cells signaling. Hy-pertension 2008;51:1010-9.19. Katz L, de Amorim M, Fiqueiroa J, Pinto eSilva J. Postpartum dexamethasone for womenwith hemolysis, elevated liver enzymes, and lowplatelets (HELLP) syndrome: a double-blind,placebo-controlled, randomized clinical trial.Am J Obstet Gynecol 2008;198:1-8.20. Martin JJ, Owens M, Keiser S, et al. Stan-dardized Mississippi Protocol treatment of 190patients with HELLP syndrome: slowing dis-ease progression and preventing new majormaternal morbidity. Hypertens Pregnancy2012;31:79-90.21. Fonseca J, Mendez F, Catano C, Arias F.Dexamethasone treatment does not improvethe outcome of women with HELLP syndrome:a double-blind, placebo-controlled randomizedclinical trial. Am J Obstet Gynecol 2005;193:1591-8.22. Martin JJ, Rinehart B, May W, Magann E,Terrone D, Blake P. The spectrum of severepreeclampsia: comparative analysis by HELLPsyndrome classification. Am J Obstet Gynecol1999;180:1373-84.23. Wallukat G, Homuth V, Fischer T, et al. Pa-tients with preeclampsia develop agonistic au-toantibodies against the angiotensin AT1 re-
ceptor. J Clin Invest 1999;103:945-52.
gy MAY 2013

24. Herse F, Verlohren S, Wenzel K, et al. Prev-alence of agonistic autoantibodies against theangiotensin II type 1 receptor and soluble fms-like tyrosine kinase 1 in a gestational age-matched case study. Hypertension 2009;53:393-8.25. Verlohren S, Herraiz I, Lapaire O, et al. ThesFlt-1/PlGF ratio in different types of hyperten-sive pregnancy disorders and its prognostic po-tential in preeclamptic patients. Am J ObstetGynecol 2012;206:58.e18.26. Martin JJ, Blake P, Lowery S, Perry KJ, FilesJ, Morrison J. Pregnancy complicated by pre-eclampsia-eclampsia with HELLP syndrome:how rapid is postpartum recovery? Obstet Gy-necol 1990;76:737-41.27. Altman D, Carroli G, Duley L, et al. Dowomen with pre-eclampsia, and their babies,benefit from magnesium sulphate? The MagpieTrial: a randomised placebo-controlled trial.Lancet 2002;359:1877-90.28. Euser A, Cipolla M. Resistance artery vaso-dilation to magnesium sulfate during pregnancyand the postpartum state. Am J Physiol HeartCirc Physiol 2005;288:H1521-5.29. Martin JJ, Thigpen B, Moore R, Rose C,Cushman J, May W. Stroke and severe pre-eclampsia and eclampsia: a paradigm shift fo-cusing on systolic blood pressure. Obstet Gy-necol 2005;105:246-54.30. Llinas M, Wallukat G, Mueller D, et al. Ago-nistic autoantibodies to the AT1 receptor in a ratmodel of preeclampsia induced by chronic re-ductions in uterine perfusion pressure (RUPP).Hypertension 2005;46:883.31. Murphy S, LaMarca B, Cockrell K, GrangerJ. Role of endothelin in mediating soluble fms-like tyrosine kinase 1 induced hypertension inpregnant rats. Hypertension 2010;55:394-8.32. Aggarwal P, Chandel N, Jain V, Jha V. Therelationship between circulating endothelin-1,soluble fms-like tyrosine kinase-1 and solubleendoglin in preeclampsia. J Hum Hypertens2012;26:236-41.33. Tam Tam K, Lamarca B, Arany M, et al. Roleof reactive oxygen species during hypertensionin response to chronic antiangiogenic factor(sFlt-1) excess in pregnant rats. Am J Hypertens2011;24:110-3.34. Levine R, Maynard S, Qian C, et al. Circu-lating angiogenic factors and risk of preeclamp-sia. N Engl J Med 2004;350:672-83.35. George E, Granger J. Linking placental isch-emia and hypertension in preeclampsia: role ofendothelin 1. Hypertension 2012;60:507-11.36. LaMarca B, Parrish M, Ray L, et al. Hyper-tension in response to autoantibodies to the an-giotensin II type I receptor (AT1-AA) in pregnantrats: a role of endothelin-1. Hypertension 2009;54:905-9.37. Lamarca B. Endothelial dysfunction. An im-portant mediator in the pathophysiology ofhypertension during pre-eclampsia. MinervaGinecol 2012;64:309-20.38. LaMarca B, Brewer J, Wallace K. IL-6 induced
pathophysiology during pre-eclampsia: potential


therapeutic role for magnesium sulfate? IntJ Interferon Cytokine Mediator Res 2011;3:59-64.39. Meekins J, McLaughlin P, West D, Mc-Fadyen I, Johnson P. Endothelial cell activationby tumour necrosis factor-alpha (TNF-alpha)and the development of pre-eclampsia. ClinExp Immunol 1994;98:110-4.40. Taylor R, Varma M, Teng N, Roberts J.Women with preeclampsia have higher plasma
endothelin levels than women with normal preg-
nancies. J Clin Endocrinol Metab 1990;71:1675-7.41. Khedun S, Naicker T, Moodley J. Endothe-lin-1 activity in pregnancy. J Obstet Gynaecol2002;22:590-3.42. Giuliano C, Smalligan R, Mitchon G, ChuaM. Role of dexamethasone in the preventionof migraine recurrence in the acute care set-ting: a review. Postgrad Med 2012;124:
110-5.
MAY 2013 Americ

43. Wang W, Wang Y, Chen X, et al. Dexameth-asone attenuates development of monocrota-line-induced pulmonary arterial hypertension.Mol Biol Rep 2011;38:3277-84.44. Ekerbicer N, Inan S, Tarakc F, Barut T, Gur-pinar T, Ozbek M. Effects of acute treatmentwith dexamethasone on hemodynamic and his-topathological changes in rats. Biotech His-
tochem 2012;87:385-96.

wallace 2013

Documents

tam tam

soluble fmnase sflt

soluble endoglin

thatthese factors

soluble fms

fetal medicine

division of maternal

hellp syndrome