Accepted Manuscript

Title: Maternal clinical disease characteristics and maternal

and neonatal outcomes in twin and singleton pregnancies with
severe preeclampsia

Author: Katherine A. Connolly Stephanie H. Factor Chloe S.

Getrajdman Catherine A. Bigelow Andrea S. Weintraub
Joanne L. Stone

PII: S0301-2115(15)00430-3
DOI: http://dx.doi.org/doi:10.1016/j.ejogrb.2015.11.031
Reference: EURO 9207

To appear in: EURO

Received date: 24-9-2015

Revised date: 3-11-2015
Accepted date: 17-11-2015

Please cite this article as: Connolly KA, Factor SH, Getrajdman CS, Bigelow CA,
Weintraub AS, Stone JL, Maternal clinical disease characteristics and maternal and
neonatal outcomes in twin and singleton pregnancies with severe preeclampsia,
European Journal of Obstetrics and Gynecology and Reproductive Biology (2015),
http://dx.doi.org/10.1016/j.ejogrb.2015.11.031

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 Katherine A. Connolly -- 1

TITLE: Maternal clinical disease characteristics and maternal and neonatal outcomes in twin and

singleton pregnancies with severe preeclampsia

AUTHORS:

Katherine A. Connolly, MD1

t
Stephanie H. Factor, MD MPH1,2

ip
Ms. Chloe S. Getrajdman1

cr
Catherine A. Bigelow, MD1,4

us
Andrea S. Weintraub, MD3

Joanne L. Stone, MD1

SITE OF RESEARCH: an
New York, NY, USA
M
INSTITUTIONS:
1
Icahn School of Medicine at Mount Sinai, Department of Obstetrics, Gynecology and
ed

Reproductive Science, New York, NY

2
Icahn School of Medicine at Mount Sinai, Department of Medicine, New York, NY
pt

3
Icahn School of Medicine at Mount Sinai, Department of Pediatrics, New York, NY
ce

4
Brigham and Womens Hospital, Department of Obstetrics and Gynecology, Boston, MA
Ac

CORRESPONDING AUTHOR:

Joanne Stone, MD
5 East 98th Street, 2nd Floor
New York, New York 10029
Phone: 212-241-5681
Fax: 212-348-7438
Email: joanne.stone@mssm.edu

Page 1 of 28
 Katherine A. Connolly -- 2

CONDENSATION:

Despite clinician concerns, mothers and neonates of twin pregnancies complicated by severe

preeclampsia do not appear to have greater morbidity and mortality when compared to mothers

and neonates of singleton pregnancies, and, in fact, singletons have a higher prevalence of

t
ip
headache, and, among those presenting 34 weeks, higher blood pressures.

cr
us
an
M
ed
pt
ce
Ac

Page 2 of 28
 Katherine A. Connolly -- 3

ABSTRACT

Objective:

Based on anecdotal observations, there is concern that severe preeclampsia leads to

t
greater morbidity and mortality for mothers and neonates of twin pregnancies than for mothers

ip
and neonates of singleton pregnancies. Because few studies have been done, this study

cr
compared maternal disease characteristics and maternal/neonatal clinical outcomes of twin and

us
singleton pregnancies complicated by severe preeclampsia.

Study Design:
an
An historical cohort study of patients hospitalized at the Mount Sinai Hospital in New

York City, NY, USA, from 2006-2010, compared 63 twin and 339 singleton pregnancies
M
complicated by severe preeclampsia via chart review. Women were analyzed in two groups:

hospitalized 34 weeks gestational age (GA) and hospitalized >34 weeks GA. Univariable
ed

analysis (using Chi-square test, Fishers Exact test, Students T-Test, or Wilcoxon Rank-Sum

test, as appropriate) then multivariable analysis (using multivariable linear regression or

pt

multivariable logistic regression, as appropriate) compared maternal disease characteristics and

ce

maternal/neonatal clinical outcomes in twin and singleton pregnancies.

Results:
Ac

Women with twins were older [mean age 34.9 years (standard deviation (SD) 7.9 years)

vs. 29.4 years (SD 7.4 years), P-value <.001] and women with singletons had a higher prevalence

of chronic hypertension (21% vs. 8%, P=.02) and higher prevalence of history of preeclampsia

(13% vs. 2%, P=.006).

Page 3 of 28
 Katherine A. Connolly -- 4

Women with twins were admitted for severe preeclampsia at an earlier gestational age

(GA) [median twin 34.9 weeks GA (interquartile range, IQR, 32.7, 36.1) vs. median singleton

37.1 weeks GA (IQR 35.0, 38.9), P < .001]. Among women presenting 34 weeks GA (27

twins; 108 singletons), women with singletons had a higher mean systolic blood pressure (BP)

t
(181.1 vs. 163.5, P < .001), higher mean diastolic BP (108.4 vs. 100.1, P = .002), and higher

ip
prevalence of headache (56% vs. 30%, P = .02). Among women presenting >34 weeks GA (36

cr
twins; 231 singletons), women with singletons had a higher prevalence of headache (54% vs.

us
28%, P = .004).

Conclusion:
an
Mothers and neonates of twin pregnancies complicated by severe preeclampsia do not

appear to have greater morbidity and mortality compared to mothers and neonates of singleton
M
pregnancies complicated by severe preeclampsia.
ed

KEY WORDS/PHRASES

severe preeclampsia; twins; historical cohort study; headache; blood pressure

pt
ce
Ac

Page 4 of 28
 Katherine A. Connolly -- 5

INTRODUCTION

Preeclampsia complicates 5-8% of pregnancies(1) and places pregnant women and their fetuses

at increased risk of morbidity and mortality. Management of pregnancies with preeclampsia is

t
based on the severity of disease. Women with twin pregnancies are 2-3 times more likely to

ip
develop preeclampsia than women with singleton pregnancies.(1) As the incidence of twin

cr
gestation increases, the impact of this association is increasingly important.(2) Based on our

us
anecdotal personal interactions with and observations of practicing obstetricians, there is concern

that severe preeclampsia leads to greater morbidity and mortality for mothers and neonates of
an
twin pregnancies than for mothers and neonates of singleton pregnancies. However, there are

few studies comparing the clinical characteristics of preeclampsia disease, and maternal and
M
neonatal outcomes in twin vs. singleton pregnancies.

Sibai et al found that twin gestations complicated by preeclampsia have a higher rate of
ed

preterm delivery when compared to singletons.(3) However, it is not known whether this was

the result of increased incidence of preterm labor in twins, as 75% of patients were diagnosed
pt

with preeclampsia intrapartum or postpartum. While the study found higher rates of
ce

preeclampsia in twin gestations, clinical characteristics of preeclampsia disease in twins

compared to singletons were not evaluated.

Ac

Henry et al(4) compared clinical characteristics of preeclampsia disease and maternal and

neonatal outcomes of singleton and twin pregnancies with severe preeclampsia at the time of

delivery, and found similar maternal and neonatal outcomes. However, these findings are of

limited utility in predicting patients clinical courses because the study participants were

evaluated at the time of delivery rather than at the time of diagnosis.

Page 5 of 28
 Katherine A. Connolly -- 6

We performed an historical cohort study to test the following null hypotheses: in

pregnancies complicated by severe preeclampsia, twin pregnancies versus singleton pregnancies

are not associated with differences in maternal clinical disease characteristics, and, in

pregnancies complicated by severe preeclampsia, twin pregnancies versus singleton pregnancies

t
are not associated with differences in maternal or neonatal outcomes.

ip
cr
MATERIALS AND METHODS

us
An institutional database was queried to identify all twin and singleton pregnancies complicated

by severe preeclampsia delivered at The Mount Sinai Hospital from 2006-2010. Severe
an
preeclampsia was defined as having one or more of the following ACOG criteria: systolic blood

pressure (BP) 160 or diastolic BP 110 on two instances at least 6 hours apart; symptoms of
M
severe preeclampsia (severe headache, visual changes, epigastric pain); abnormal lab values

(transaminase values > twice normal, platelets <100,000, >5g proteinuria in 24 hours); oliguria
ed

(<500mL in 24 hours); intrauterine growth restriction (IUGR); or pulmonary edema.(1) The

medical records of all potential subjects were reviewed to confirm the diagnosis of severe
pt

preeclampsia.1 Pregnant patients were excluded if they had higher order multiple gestation,
ce

presented with intrauterine fetal demise, or had incomplete admission records.

The following clinical information was collected retrospectively from the medical records
Ac

of pregnant study subjects: demographic data (age, race, insurance, marital status); social habits

(cigarette smoking, alcohol use, illicit drug use); past medical history/medical comorbidities

(chronic hypertension, asthma, history of sexually transmitted infections, HIV infection,

hyperthyroidism, hypothyroidism, lupus, kidney disease); past obstetrical history (gravidity,

parity, gestational diabetes, preeclampsia, preterm delivery); comorbidities of current pregnancy

Page 6 of 28
 Katherine A. Connolly -- 7

(gestational hypertension prior to diagnosis of preeclampsia, gestational diabetes,

oligohydramnios, final body mass index, admission in labor or with premature rupture of

membranes or preterm premature rupture of membranes); characteristics of current preeclampsia

disease (highest blood pressures, oliguria, visual change, pulmonary edema, peripheral edema,

t
epigastric right upper quadrant pain, seizure, headache, development of the syndrome of

ip
hemolysis, elevated liver enzymes, and low platelet count (HELLP syndrome), eclampsia,

cr
stroke); and current laboratory studies (protein in the urine, hematuria, lactate dehydrogenase,

us
aspartate aminotransferase, alanine aminotransferase, partial thromboplastin time, prothrombin

time, creatinine, uric acid, platelet count, and fibrinogen). Maternal management choices,
an
including treatment with antihypertensive medication, treatment with magnesium sulfate, and

days from admission to delivery were also noted.

M
The following clinical information was collected retrospectively from the medical records

of newborns of the study mothers: small for gestational age (birth weight < 10th percentile), need
ed

for neonatal intensive care unit admission, need for ventilator support, and neonatal mortality.

Statistical analysis
pt

At The Mount Sinai Hospital, patients diagnosed with severe preeclampsia > 34 weeks GA are
ce

delivered at the time of presentation, whereas patients diagnosed at or prior to 34 weeks may be

candidates for expectant management until 34 weeks.(5) Therefore, the analysis was stratified
Ac

into a 34 weeks GA group and > 34 week GA group.

Clinical characteristics of disease and maternal and neonatal outcomes were then

compared in univariable analysis using Chi-square, Fishers Exact, T-tests or Wilcoxon Rank

Sum tests, as appropriate. Clinical characteristics and maternal/neonatal outcomes that were

significant in univariable analyses were evaluated further by multivariable analysis; continuous

Page 7 of 28
 Katherine A. Connolly -- 8

variables were evaluated with multivariable linear regression, and categorical variables were

evaluated with multivariable logistic regression.

The multivariable analysis included building a multivariable model for each of the

clinical characteristics of preeclampsia disease and maternal/neonatal outcomes found to be

t
significantly different between twins and singletons. Specifically, building a multivariable

ip
model used the clinical characteristics of preeclampsia disease or maternal/neonatal outcomes as

cr
the outcome variable, twin status as the main exposure, GA at presentation as a necessary

us
covariate, and the differences in maternal demographic data, past medical history/medical

comorbidities, prior pregnancy history, and characteristics of current pregnancy that differed
an
between the twins and singleton pregnancies (as described above) as possible confounders.

Forward, backward, and stepwise regression was then used to identify the most parsimonious
M
model. When the final multivariable model indicated that twin status was statistically associated

with the outcome (P-value .05), this was noted and presented in the Results.
ed

SAS Version 9.3 (Cary, NC) was used for the statistical analysis. A P-value of .05 was

considered significant. Missing data was assumed to be missing at random; statistical analyses
pt

used only available data. The study was approved by the Institutional Review Board of The
ce

Icahn School of Medicine at Mount Sinai.

Ac

RESULTS

The selection of the study population is presented in Figure 1. There were 29,025 singleton and

twin births at The Mount Sinai Hospital during 2006-2010; 27,886 were singleton and 1139 were

twin births. Of these, there were 525 singleton and 63 twin births in which the mother presented

with severe preeclampsia. In our study population, the upper limit of gestation for a twin

Page 8 of 28
 Katherine A. Connolly -- 9

pregnancy presenting with severe preeclampsia was 38 1/7 weeks; therefore, to ensure

comparability of twin and singleton groups, all singleton pregnancies with severe preeclampsia

presenting after 38 1/7 weeks were excluded. Of the 525 women with singleton pregnancies,

339 presented at 38 1/7 weeks GA and were therefore eligible for inclusion in this study.

t
(Figure 1).

ip
Compared to women with singleton gestations, women with twin pregnancies were older,

cr
less like to be Black and/or Latina, and more likely to have private insurance, to be married, to

us
be nulliparous, and to be hypothyroid (Table 1). Compared to women with twin gestations,

women with singleton pregnancies were more likely to have chronic hypertension, a history of
an
preeclampsia, and a history of preterm delivery (Table 1). Women with twins were more likely

to be admitted for severe preeclampsia at an earlier GA than women with singletons (Figures 2A
M
and 2B).

There were 135 women who presented with severe preeclampsia at 34 weeks GA
ed

(Table 2). In univariable and then multivariable analysis, the mean highest systolic and diastolic

BPs were lower among women with twin pregnancies than among women with singleton
pt

pregnancies (Table 2 and 3). Women with twin pregnancies were less likely to have a headache
ce

than women with singleton pregnancies (Table 2 and Table 4).

There were 267 women who presented with severe preeclampsia at 34 weeks gestation
Ac

(Table 2). In univariable then multivariable analysis, women with twin pregnancies were less

likely to have a headache (Table 2 and Table 4).

To determine if singleton and twin gestations complicated by severe preeclampsia were

managed differently, clinical management choices were examined based on the GA at

presentation. Consistent with our finding that women with singleton pregnancies who presented

Page 9 of 28
 Katherine A. Connolly -- 10

at 34 weeks gestation with severe preeclampsia had higher systolic and diastolic BP than

women with twin pregnancies, a higher percentage of women with singleton pregnancies who

presented at 34 weeks gestation were treated with antihypertensive medication (Table 5).

There were no differences in the median days from admission to delivery for women singleton

t
vs. twin pregnancies complicated by severe preeclampsia.

ip
cr
COMMENT

us
In this historical cohort study of singleton and twin gestations complicated by severe

preeclampsia, mothers with twin gestations present earlier than mothers with a singleton
an
pregnancy. Mothers and neonates of twin pregnancies complicated by severe preeclampsia do

not appear to have greater morbidity and mortality when compared to mothers and neonates of
M
singleton pregnancies complicated by severe preeclampsia. In fact, singleton pregnancies with

severe preeclampsia presenting at 34 weeks gestation had the mean highest systolic and
ed

diastolic BPs and a higher prevalence of headaches. When comparing singleton and twin

pregnancies complicated by severe preeclampsia presenting at > 34 weeks gestation, singleton

pt

pregnancies also had a higher prevalence of headaches. The clinical significance of these
ce

differences in BP and headache are not obvious, as no difference in the prevalence of seizures or

stroke was found.

Ac

Our results are in agreement with the finding by Sibai et al(3) that women with twin

gestations complicated by severe preeclampsia are more likely to be delivered preterm than

singletons. Although in our study population, women with twin gestation complicated by

preeclampsia presented earlier to the hospital, we did not find any differences in presentation

with premature labor based on twin status nor did we find any differences in latency from

Page 10 of 28
 Katherine A. Connolly -- 11

admission to delivery based on twin status. Our data therefore suggests that twins with severe

preeclampsia deliver earlier because they present earlier.

Henry et al.(4) showed that twin gestations are more likely to develop severe

preeclampsia earlier than singletons, but disease manifestations and severity appeared to be

t
similar. In our study population, unlike the population studied by Henry et al, there were no

ip
significant differences in serum laboratory abnormalities, prevalence of HELLP syndrome, or

cr
placental abruption in twin vs. singleton pregnancies complicated by preeclampsia. However,

us
we did discern significant differences in systolic and diastolic BP and prevalence of headache

between the two groups, which are disease manifestations not investigated by Henry et al. In
an
both studies, mothers of singletons who developed severe preeclampsia were significantly more

likely to have a personal history of chronic hypertension. This suggests that while twin
M
pregnancies have an increased incidence of preeclampsia, mothers of singletons who develop the

disease may have preexisting vascular disease. Both the Henry et al(4) study and this current
ed

study suggest that the clinical characteristics of severe preeclampsia disease and the

maternal/neonatal outcomes from severe preeclampsia are similar for both twins and singletons.
pt

It is possible that the observed differences may be due to differential practices between
ce

twin and singletons; caregivers may be more expeditious and deliver twins faster than singletons

leading to more adverse effects in the case of singletons. We did not observe a difference in the
Ac

time from admission to delivery between twins and singletons when counted in days. Perhaps

days are too gross a time course by which to measure time from admission to delivery.

Ours is the first study to document higher BPs in women with singleton pregnancies with

severe preeclampsia compared to women with twin pregnancies with severe preeclampsia. The

difference in BP remained even after evaluation for possible confounding by history of

Page 11 of 28
 Katherine A. Connolly -- 12

preeclampsia and history of chronic hypertension. It is possible that women pregnant with

singletons who develop severe preeclampsia have increased underlying vascular endothelial

damage, putting them at risk for the disease. Biologically, it is also possible that a larger

placenta in a twin pregnancy may decrease systemic vascular resistance and thus account for the

t
lower mean BP seen in twin gestations who present at or before 34 weeks gestation. However,

ip
previous studies have suggested that increased placental mass is associated with an increased risk

cr
of preeclampsia in twins.(6) Future research may identify the etiology of the observed

us
differences in BPs and may determine if the differences are clinically important.

This is also the first study to document a higher prevalence of headache in singleton
an
pregnancies with severe preeclampsia than twin pregnancies with preeclampsia. The cause of

headache in women with preeclampsia is poorly understood.(7) Some have suggested that
M
headache in women with preeclampsia is associated with an abnormality in cerebral perfusion

pressure(7) while others have suggested vascular endothelial dysfunction.(8) The reason for the
ed

higher prevalence of headache among singleton pregnancies may be due to differences in the

underlying pathophysiology of preeclampsia disease in twins and singletons or a difference in

pt

underlying maternal vascular endothelial damage. Future research may identify the etiology of
ce

the differences in headache.

Our study has several limitations including the retrospective nature of our review. We
Ac

were unable to adjust for history of in vitro fertilization, which is a known risk factor for

preeclampsia,(9) other assisted methods of reproduction, or chorionicity, as this information was

not available in the admission records. As a single center study, our results may not be fully

generalizable to all patient population. There were differences in management which cannot be

controlled in observational studies and were difficult to determine from chart review. We used

Page 12 of 28
 Katherine A. Connolly -- 13

GA at presentation as a surrogate for GA of diagnosis of severe preeclampsia. The exact onset

of severe preeclampsia in the outpatient setting cannot be determined without continuous daily

monitoring which is not standard. It is possible that because twin gestations are monitored more

frequently at an earlier GA, preeclampsia in these pregnancies may have been diagnosed earlier.

t
However, hospital admission does not prevent disease progression and thus earlier admission

ip
should not impact upon the disease characteristics we observed. A related limitation is that

cr
women were only included in the study once they were admitted. Thus, we were unable to

us
evaluate disease characteristics which occurred prior to hospitalization.

This study used the working ACOG definition of severe preeclampsia at the time these
an
patients were seen.(1) This definition of preeclampsia has been modified since that time.(10)

The new definition of preeclampsia eliminates both IUGR and >5g proteinuria in 24 hours as
M
defining features.(10) If the current definition were used, all but two patients in our study

population would be diagnosed with severe preeclampsia; one was the mother of twins who
ed

presented before 34 weeks, and the other was the mother of a singleton who presented after 34

weeks. Therefore, it is unlikely that the most current definition of severe preeclampsia would
pt

change the findings of this study.

ce

This study suggests that although severe preeclampsia is more prevalent among twin

gestations, mothers and neonates of twin pregnancies are not at increased risk of morbidity and
Ac

mortality when compared to mothers and neonates from singleton pregnancies. This finding

should hopefully allay some clinician concerns about more severe disease in twins. With the

increasing incidence of twin gestations, further research to elucidate possible differences in the

pathogenesis of preeclampsia in twins and singletons is needed.

Page 13 of 28
 Katherine A. Connolly -- 14

ACKNOWLEDGEMENTS

The authors wish to thank Ms. Nicole Cornet and Ms. Ariana Mills for their help with

data collection for this study.

t
ip
cr
us
an
M
ed
pt
ce
Ac

Page 14 of 28
 Katherine A. Connolly -- 15

REFERENCES

1. Bulletins--Obstetrics ACoP. ACOG practice bulletin. Diagnosis and management of

preeclampsia and eclampsia. Number 33, January 2002. Obstet Gynecol. 2002;99(1):159-67.
2. Russell RB, Petrini JR, Damus K, Mattison DR, Schwarz RH. The changing
epidemiology of multiple births in the United States. Obstet Gynecol. 2003;101(1):129-35.

t
3. Sibai BM, Hauth J, Caritis S, Lindheimer MD, MacPherson C, Klebanoff M, et al.

ip
Hypertensive disorders in twin versus singleton gestations. National Institute of Child Health and
Human Development Network of Maternal-Fetal Medicine Units. Am J Obstet Gynecol.

cr
2000;182(4):938-42.
4. Henry DE, McElrath TF, Smith NA. Preterm severe preeclampsia in singleton and twin
pregnancies. J Perinatol. 2013;33(2):94-7.

us
5. Sibai BM. Evaluation and management of severe preeclampsia before 34 weeks'
gestation. Am J Obstet Gynecol. 2011;205(3):191-8.
6. Bdolah Y, Lam C, Rajakumar A, Shivalingappa V, Mutter W, Sachs BP, et al. Twin
pregnancy and the risk of preeclampsia: bigger placenta or relative ischemia? Am J Obstet
Gynecol. 2008;198(4):428 e1-6.
7.
an
Belfort MA, Saade GR, Grunewald C, Dildy GA, Abedejos P, Herd JA, et al. Association
of cerebral perfusion pressure with headache in women with pre-eclampsia. British journal of
obstetrics and gynaecology. 1999;106(8):814-21.
M
8. Block HS, Biller J. Neurology of pregnancy. Handbook of clinical neurology.
2014;121:1595-622.
9. Jackson RA, Gibson KA, Wu YW, Croughan MS. Perinatal outcomes in singletons
ed

following in vitro fertilization: a meta-analysis. Obstet Gynecol. 2004;103(3):551-63.

10. American College of O, Gynecologists, Task Force on Hypertension in P. Hypertension
in pregnancy. Report of the American College of Obstetricians and Gynecologists' Task Force on
Hypertension in Pregnancy. Obstet Gynecol. 2013;122(5):1122-31.
pt

Legends for Figures and Tables

ce
Ac

Page 15 of 28
 t
ip
Katherine A. Connolly -- 17

cr
Table 1. Maternal characteristics of women with singleton and twins pregnancies presenting with severe preeclampsia to Mount Sinai

Hospital, 2006-2010 (N=402).5

us
Twins Singletons

Characteristic (N=63)(%) (N=339)(%) P-value

an
Demographics

M
Maternal mean age (SD) 34.9 (7.9) 30.2 (7.4) <.0012

Race/Ethnicity

d
Latino/Black 17 (28) 219 (65) <.0011

Caucasian/Other

Type of insurance
te 43 (72) 119 (35)
ep
Medicaid 14 (22) 179 (54) <.0011
c

Private 49 (78) 152 (46)

Ac

Marital Status

Single 16 (28) 179 (54) <.0011

Married 42 (72) 151 (46)

Cigarette Smoker 1 (2) 22 (7) .233

Page 16 of 28
 t
ip
Katherine A. Connolly -- 18

cr
Alcohol 1 (2) 11 (3) 1.003

Illicit drugs 1 (2) 10 (3) 1.003

us
Past Medical History

.0021

an
Chronic hypertension 5 (8) 89 (26)

Asthma 9 (15) 62 (18) .461

M
Pre-gestational diabetes mellitus 1 (2) 19 (6) .343

History of sexually transmitted infections 11 (18) 65 (19) .771

d
HIV infection 0 1 (0.3) 1.003

Hypothyroid

Hyperthyroid
te 7 (11)

1(2)
11 (3)

3 (1)
.013

.49
ep
Lupus 0 6 (2) .603

.483
c

Renal Disease 1 (2) 14 (4)

Ac

Prior Pregnancy History

History of gestational diabetes 6 (10) 42 (12) .511

History of preeclampsia 1 (2) 55 (16) <.0013

History of preterm delivery 2 (3) 63 (19) .0013

Page 17 of 28
 t
ip
Katherine A. Connolly -- 19

cr
Nulliparous 47 (75) 190 (56) .0061

Characteristics of Study Pregnancy

us
GA at presentation for hospitalization

Median (IQR) 34.9 (32.7, 36.1) 35.7 (33.4, 37.1) .0074

an
Gestational diabetes during study pregnancy 0 0 ND

M
Oligohydramnios 0 1 (0.3) 1.003

Final BMI

d
Median (IQR) 31.0 (28.7, 35.9) 32.1 (27.8, 38.1) .554

1
Calculated using Chi-Square
te
ep
2
Calculated using Students T-test
3
Calculated using Fishers Exact
c
Ac

4
Calculated using Wilcoxon rank-sum test
5
Numbers in the cells may not add up to the total N due to missing data.

ND Not defined; BMI Body mass index; IQR interquartile range

Page 18 of 28
 t
ip
Katherine A. Connolly -- 20

cr
Table 2. Univariable analysis of clinical characteristics of mothers and neonates in singleton vs. twin pregnancies complicated by

us
severe preeclampsia

Gestational age at presentation

an
34 weeks > 34 weeks
Twins Singletons Twins Singletons
(N = 27)(%) (N = 108)(%) P-value (N = 36)(%) (N = 231)(%) P-value

M
Antepartum characteristics of preeclampsia
Mean highest systolic blood pressure 163.515.7 181.119.5 <.0012 165.715.3 168.716.3 .272
Mean highest diastolic blood pressure 100.111.4 108.413.6 .0022 98.613.5 102.410.4 .112
Oliguria 2 (7) 7 (6) 1.003 1 (3) 10 (4) 1.003

d
Visual change 3 (11) 21 (19) .413 8 (22) 51 (22) .983
Pulmonary edema 0 2 (2) 1.003 0 2 (1) 1.003
Peripheral edema
te
Epigastric right upper quadrant pain
Seizure
15 (56)
4 (15)
0
73 (68)
16 (15)
1 (1)
.241
1.003
1.003
19 (53)
3 (8)
0
132 (57)
23 (10)
0
.621
1.003
ep
Headache 8 (30) 60 (56) .021 10 (28) 124 (54) .0041
HELLP syndrome 12 (44) 45 (42) .791 12 (33) 57 (25) .271
Eclampsia 0 0 0 0
c

Stroke 0 1 (1) 1.003 0 0

Laboratory Studies
Ac

Proteinuria based on dipstick 25 (93) 104 (97) .263 35(97) 219 (96) 1.003
Highest LDH, IQR (in U/L) 408.5 (314,455) 383 (309, 487) .914 326 (266, 414) 336 (285, 437) .464
Abnormal ALT and/or AST 11 (41) 48 (44) .731 12 (34) 62 (27) .371
PTT > 35 seconds 4 (15) 8 (7) .253 3 (9) 30 (13) .593
PT > 15.5 seconds 3 (12) 5 (5) .203 0 12 (5) .373
Fibrinogen < 150 mg/dL 1 (4) 1 (1) .363 0 0
Highest creatinine, IQR (mg/dL) 0.8 (0.5, 1.0) 0.7 (0.6, 0.9) .354 0.8 (0.5, 0.9) 0.7 ( 0.6, 0.8) .294
Highest uric acid (mg/dL) 7.0 (1.6) 6.3 (1.5) .132 6.4 ( 1.8) 6.1 ( 1.6) .282

Page 19 of 28
 t
ip
Katherine A. Connolly -- 21

cr
Platelets < 150,000 /L 13 (48) 66 (61) .221 18 (50) 164 (71) .011
Delivery characteristics
Admitted in labor or with PROM 2 (7) 2 (3) .263 6 (17) 24 (10) .271

us
PPROM 1 (4) 0 .203 0 0
Neonatal characteristics
(N=54) (N=108) (N=72) (N=231)
1
.191

an
SGA 20(37) 27(25) .11 12 (17) 25 (11)
NICU admission 43 (80) 95 (88) .161 17 (24) 53 (23) .911
Assisted ventilation 15 (28) 37 (34) .401 0 3 (1) 1.003
Neonatal mortality 2 (4) 4 (4) 1.003 1 (1) 1 (0.4) .423

M
1
Univariable analysis calculated using Chi-Square

d
2
Univariable analysis calculated using Students T-test
3

4
te
Univariable analysis calculated using Fishers Exact

Univariable analysis calculated using Wilcoxon Rank Sum

ep
SDStandard deviation; IQR interquartile range; HELLP syndrome hemolysis, elevated liver enzymes, and low platelets; LDH
c

lactate dehydrogenase; ALT alanine aminotransferase; AST -- aspartate aminotransferase; PTT partial thromboplastin time; PT
Ac

Prothrombin time; PROM premature rupture of the membranes; PPROM preterm premature rupture of the membranes; SGA

small for gestational age (weight below 10th percentile for gestational age); NICUneonatal intensive care unit

Page 20 of 28
 t
ip
Katherine A. Connolly -- 22

cr
Table 3. Multivariable linear regression analysis of clinical characteristics of mothers and neonates in singleton vs. twin pregnancies

us
complicated by severe preeclampsia who presented 34 weeks gestational age (GA)1

Highest systolic blood pressure (Model 1)2 Highest diastolic blood pressure (Model 2)3
Parameter Standard error P-value Parameter Standard error P-value

an
estimate estimate
Twin vs. singleton -14.85 4.00 <.001 -8.14 2.83 .005
GA at presentation -0.76 0.58 .19 -0.52 0.42 .22

M
Maternal Age 0.75 0.23 .001
Medicaid vs. 10.37 3.57 .004
private insurance
Chronic 7.96 3.57 .03

d
hypertension

1
te
GA at presentation was included in both models as a necessary covariate. For each model maternal age, race/ethnicity, insurance
ep
type, marital status, chronic hypertension, hypothyroidism, history of preeclampsia, history of preterm delivery, and nulliparity were

evaluated as possible confounders. The results show the most parsimonious model.
c

2
Of 135 pregnancies of singletons and twins who presented 34 weeks GA, 134 had complete data for all of the variables included in
Ac

Model 1 and were therefore included in Model 1.

3
Of 135 pregnancies of singletons and twins who presented 34 weeks GA, 135 had complete data for all of the variables included in

Model 2 and were included in Model 2.

Page 21 of 28
 t
ip
Katherine A. Connolly -- 23

cr
Table 4. Multivariable logistic regression analysis of clinical characteristics of mothers and neonates in singleton vs. twin pregnancies

us
complicated by severe preeclampsia 1

Headache among pregnancies that presented at 34 Headache among pregnancies that presented at > 34
weeks gestational age (Model 3)2 weeks gestational age (Model 4)3

an
OR (95%CI) P-value OR(95%CI) P-value
Twin vs. singleton 0.34 (0.14, 0.85) .02 0.31 (0.14, 0.68) .004
GA at presentation 0.97 (0.85, 1.10) .61 0.90 (0.72, 1.23) .35

M
1
GA at presentation was included in both models as a necessary covariate. For each model, maternal age, race/ethnicity, insurance

d
type, marital status, chronic hypertension, hypothyroidism, history of preeclampsia, history of preterm delivery, and nulliparity were

2 te
evaluated as possible confounders. The results show the most parsimonious model.

Of 135 pregnancies of singletons and twins who presented 34 weeks GA, 135 had complete data for all of the variables included in
ep
Model 3 and were therefore included in Model 3.
3
Of 267 pregnancies of singletons and twins who presented > 34 weeks GA, 267 had complete data for all of the variables included in
c
Ac

Model 4 and were therefore included in Model 4.

GA gestational age

Page 22 of 28
 t
ip
Katherine A. Connolly -- 25

cr
Table 5. Gestational age at admission and clinical management in singleton vs. twin gestations complicated by severe preeclampsia

Gestational age at presentation to Mount Sinai Hospital

us
34 weeks >34 weeks
Twins Singletons Twins Singletons
(N = 27) (N = 108) P-value (N = 36) (N = 231) P-value
.224 .0054,5

an
Median Admission GA (IQR) 32.4 (30.3, 33.0) 31.7 (29.5, 33.2) 35.7 (35.1, 36.6) 36.6 (35.6, 37.4)
4
Median days from admission to 2.0 (0, 7.0) 2.0 (0, 3.0) .13 0 (0, 1.0) 1.0 (0, 1.0) .064
delivery (IQR)
Median GA at delivery (IQR) 33.0 (31.4, 33.4) 32.1 (30.3, 33.6) .224 35.9 (35.1, 36.7) 36.9 (35.7, 37.4) .0024,5

M
Treated with antihypertensives 11 (41) 82 (76) <.0011,6 14 (39) 107 (47) .391
Treated with magnesium sulfate 23 (85) 101 (94) .233 30 (83) 213 (92) .081

d
1
Chi-Square
2

3
Students T-test

Fishers Exact
te
ep
4
Calculated using Wilcoxon Rank Sum
5
c

Twin status was significantly associated with this characteristic (P .05) in multivariable linear regression, adjusting for gestational
Ac

age at presentation, and after adjusting for possible confounding by maternal age, race/ethnicity, insurance type, marital status, chronic

hypertension, hypothyroid, history of preeclampsia, history of preterm delivery, and nulliparity.

6
Twin status was significantly associated with this characteristic (P .05) in multivariable logistic regression, adjusting for gestational

age at presentation, and after adjusting for possible confounding by maternal age, gestational age at presentation, race/ethnicity,

insurance type, marital status, chronic hypertension, hypothyroid, history of preeclampsia, history of preterm delivery, and nulliparity.

Page 23 of 28
 t
ip
Katherine A. Connolly -- 26

cr
GA gestational age; IQR interquartile range

us
an
M
d
te
c ep
Ac

Page 24 of 28
 Katherine A. Connolly -- 28

FIGURE LEGENDS

FIGURE 1.

Title: Selection of study subjects, Mount Sinai Hospital, 2006-2010

t
FIGURE 2A

ip
Title: Number of women with singleton pregnancy complicated by severe preeclampsia by

cr
gestational age of presentation, Mount Sinai Hospital, 2006 2010 (N = 339).

us
FIGURE 2B
an
Title: Number of women with twin pregnancy complicated by severe preeclampsia by gestational

age of presentation, Mount Sinai Hospital, 2006 2010 (N = 63).

M
ed
pt
ce
Ac

Page 25 of 28
Figure 1

t
ip
cr
us
an
M
d
te
p
ce
Ac

Page 26 of 28
Figure 2A

i
cr
us
an
M
ed
pt
ce
Ac

Page 27 of 28
Figure 2B

i
cr
us
an
M
ed
pt
ce
Ac

Page 28 of 28