SRI Supplement to Reproductive Sciences - Volume 25 Number 1 - March 2018 - 232A

232A

Reproductive Sciences Vol. 25, Supplement 1, March 2018

use alone, and both. Prevalence of substance use was weighted to account
for oversampling of black race, preterm birth and placental pathology
availability. MJ use was detected by self-report or cord homogenate
positive for 11-nor-delta-9-tetrahydrocannabinol-9-carboxylic acid.
Tobacco use was detected by self-report or presence of maternal serum
cotinine. Stillbirths and live-births were considered separately. The
primary outcome was the FPR (grams fetus/grams placenta). The effect
of MJ and tobacco use on FPR was estimated with multivariable linear
modeling adjusted for fetal sex and gestational age at delivery.
RESULTS: Of 1663 women with information regarding MJ use and
placental pathology available, 504 (35%) were stillbirths and 1159 (65%)
were live-births. Among live-births, 85% did not use tobacco or MJ, 1%
used MJ only, 11% used tobacco only, and 2% used both MJ and tobacco.
Fetal weights, placental weights and FPR were similar between groups in
univariable analysis. In multivariable modeling adjusting for fetal sex and
gestational age at delivery, MJ alone and tobacco alone were not associated
with significant differences in the FPR compared to women with no use.
However, the concurrent use of MJ and tobacco resulted in a lower FPR
than using MJ alone (p=0.03; Figure). Among stillbirths (n=504), 77%
did not use tobacco or MJ, 1% used MJ only, 18% used tobacco only,
and 3% used both MJ and tobacco. There were no differences in FPR
between groups in univariable comparisons or multivariable modeling.
CONCLUSION: Exposure to both MJ and tobacco was associated with
lower FPR among live-births, which may reflect acute placental injury
and poor function rather than poor placental growth. Further evaluation
of placental pathologic features may provide insight into the mechanism
of the adverse effects of tobacco and MJ on fetal growth.
*Figure(s) will be available online.

Scientific Abstracts

from non-hemorrhage patients, with median MoMs of 1.01 and 0.94
(aOR=1.02, 95% CI 0.10-10.83). All other markers showed moderate,
non-significant differences.
CONCLUSION: Both PAPP-A and AFP show promise in identifying
patients with a morbid accreta, while Estriol fails to do so. Larger studies
are needed to validate these findings and evaluate optimal cutoff values
that predict hemorrhage from accreta.
Serum Screen Marker MoM as Predictor of Hemorrhage in Patients with
Accreta
No Hemorrhage
Median (range)

Hemorrhage Median Odds
(range)
Ratio

Papp_A

1.05 (0.80-1.51)
N=79

1.34 (1.01-2.29)
N=45

1.89
(1.163.10)

1.80 (1.043.10)

hCG (First 1.07 (0.74-1.47)
Trimester) N=63

1.28 (0.91-1.74)
N=39

1.80
(0.873.70)

1.93 (0.894.20)

hCG
(Second
Trimester)

1.05 (0.82-1.37)
N=20

1.50 (0.78-2.2) N=18

1.80
(0.774.23)

1.40 (0.573.45)

MSAFP

1.16 (0.94-1.5)
N=24

1.54 (1.22-2.14)
N=21

2.80
(0.968.14)

3.40 (0.8713.38)

Estriol

0.94 (0.84-1.06)
N=20

1.01 (0.76-1.27)
N=18

1.45
(0.267.96)

1.02 (0.1010.83)

Inhibin

1.38 (0.78-2.14)
N=20

1.59 (1.05-2.31)
N=17

1.54
(0.683.45)

1.19 (0.463.12)

F-153
Serum Screens as a Predictor of Hemorrhage in Placenta Accreta.
Nicola Perlman†,1,3 Megan E Bunnell†,1,2 Daniela A Carusi*.1,3 1Brigham
& Women's Hospital, Boston, MA, United States; 2Dartmouth College,
Hanover, NH, United States; 3Harvard Medical School, Boston, MA,
United States.
INTRODUCTION: Placenta accreta, a condition of abnormally invasive
placentation, is currently diagnosed pathologically. Prenatal patients at
risk of accreta are identified by history and radiology, but the range of
hemorrhagic outcomes is wide and hard to predict. The objective of this
study is to identify components of the maternal serum screen that can
select patients most likely to hemorrhage in the setting of placenta accreta.
METHODS: Retrospective cohort study of patients with placenta accreta
on pathology for whom first (PAPP-A, hCG) or second trimester (AFP,
hCG, Inhibin, Estriol) serum screen results were available between 2006
and 2014. Results were recorded as multiples of the population median
(MoM). Using logistic regression, we compared the serum screen MoMs
of accreta patients with and without major hemorrhage at delivery or
postpartum. Maternal demographics, surgical and previa history, use of
IVF and fetal number were tested as potential confounders. Hemorrhage
was defined as estimated blood loss (EBL)≥1500cc, hematocrit drop≥10%,
blood cell transfusion, use of uterine balloon tamponade, or uterine artery
embolization.
RESULTS: We identified 300 qualifying accreta patients, of whom
137 (46%) had serum screen results available. Of these, 53 (39%) met
criteria for major hemorrhage. The Estriol MoM for accreta patients
approximated the population median of 1, with a median of 0.99 (IQR
0.82-1.23, N=38). Median MoM for AFP (1.26, IQR 1.07-1.66, N=45),
PAPP-A (1.14, IQR 0.83-1.78, N=124) and Inhibin (1.41, IQR 0.90-2.15,
N=37) were all significantly higher than the Estriol median. First and
second trimester HCG MoMs did not significantly differ from Estriol.
Accreta patients with major hemorrhage had significantly higher PAPP-A
MoMs than those without (1.34 vs. 1.05, respectively, aOR=1.80, 95%
CI 1.04-3.10), after controlling for previa and multiple gestations. MoM
of AFP also distinguished the hemorrhage (1.54) from non-hemorrhage
group (1.16), though it did not reach statistical significance (aOR=3.40,
95% CI 0.87-13.38). In contrast, Estriol failed to distinguish hemorrhage

Adjusted
Odds
Ratio

Serum
Screen

F-154
Sex Specific Associations with Placental Features at Term. Joan
Krickellas†,2 Sylvia Dygulski†,2 Hannah Leah Bromberg†,2 Serena
Chen†,2 Michael Joyce†,2 Julian Nepola†,2 Jacqueline Bush,1 Beata
Dygulska,1 Pramod Narula,1 Sanford Lederman,1 Carolyn Margaret
Salafia*.2 1New York Presbyterian Brooklyn Methodist Hospital, Brooklyn,
NY, United States; 2Placental Analytics, LLC, New Rochelle, NY, United
States.
INTRODUCTION: Placental measures are most commonly only
collected from placentas submitted for pathology examination due to
clinical evidence of complication. We have a unique cohort of consecutive
placenta pathology cohort; we present here a preliminary analysis of
the first 25% of a frequency sample of the population regarding clinical
correlations with histologic evidence of acute ascending intraamniotic
infection, which we have recently demonstrated in the same population
to be associated with a significantly increased risk of autism diagnosis.
METHODS: We extracted from electronic medical records maternal
and neonatal data as well as specific features of he gross and microscopic
placental surgical pathology examination. We explored the relationship of
the measures of the maternal and fetal response to intraamniotic infection
to other placental and newborn measures in this term population. Nonparametric tests considered p<0.05 significant.
RESULTS: Fetal inflammatory responses to intraamniotic infection were
inversely correlated with placental infarct (p<0.0001) as well as chronic
extraplacental decidual chronic inflammation (p<0.0001 in females only
males p=0.744) and chronic amnionitis (p=0.001 in females, p=0.047
in males) and fetal inflammatory responses in cord and chorionic plate
vessels (each p<0.0001). Chronic amnionitis was associated with acute
chorionitis and umbilical vasculitis only in males (p=0.041, v female,
p=0.928, p<0.0001 v females p=0.87, respectively). Fetal inflammatory
responses were not related to any fetal growth parameters or to NICU
admission in either sex.
CONCLUSION: Fetal inflammatory responses measured by placental
histopathology are clinically silent, even considering a wide range of
newborn measures, Their association with autism spectrum disorder
diagnosis requires detailed understanding of the nature of subclinical
intraamniotic infection.
Table of Contents for the Digital Edition of SRI Supplement to Reproductive Sciences - Volume 25 Number 1 - March 2018
