SRI Supplement to Reproductive Sciences - Volume 25 Number 1 - March 2018 - 116A

116A

Reproductive Sciences Vol. 25, Supplement 1, March 2018

the toll-like receptor (TLR) agonist, FSL-1. Similar experiments were
carried out using a monolayer culture model. Cell viability was assessed
at the end of each treatment.
RESULTS: Treatment of VECs cultured as a monolayer with physiological
relevant concentrations of LA induces high levels of cell apoptosis and
highly variable activation of AP-1 and NFκB. In comparison, viability
of VECs treated with LA in the Transwell system remains high (75% at
18h). Basal NFκB and AP-1 activation remains constant at 30 min, 2h
and 6h with increasing doses of both L- and D-LA. Activation of TLR2/6mediated inflammatory pathways induces moderate yet transient NFκB
activation with levels returned to basal by 2h. In contrast AP-1 is strongly
activated at 2h and 6h following FSL-1 treatment.
CONCLUSION: The transwell culture system represents a more
physiologically relevant culture method for VECs cultured in acidic
conditions. TLR2/6-mediated inflammation in VECs appears to be largely
modulated through AP-1 activation. Current studies are investigating the
effect of LA isomers on inflammatory stimulated VECs.

T-018
Recognition of Vaginal Microbiota by the Adaptive Maternal Immune
System. Denise CY Chan†,2 Yun S Lee*,2 Sadhana Bura†,2 Maria
Arianoglou†,2 TG Teoh*,1 Maria Carmen Collado*,3 Phillip R Bennett*,2
David A MacIntyre*,2 Lynne Sykes*.2 1Imperial College Healthcare NHS
Trust, London, United Kingdom; 2Imperial College London, London,
United Kingdom; 3Institute of Agrochemistry and Food Technology of
the Spanish National Research Council (IATA-CSIC), Valencia, Spain.
INTRODUCTION: Microbial driven inflammation is a common cause of
spontaneous preterm labour. Successful pregnancy depends upon unique
adaption of the maternal immune system ensuring an appropriate response
to commensals and pathogens. While relationships between specific
vaginal microbiota and risk of preterm birth have been described, the role
of the individual host immune response in shaping these relationships in
largely unknown. Not all women with high-risk microbiome signatures
(e.g. L. iners or G. vaginalis dominance) deliver preterm. In this study
we aimed to develop a technique for examining the interactions between
vaginal microbiota and the host immune response, and to determine if
these interactions are altered during pregnancy.
METHODS: Cervicovaginal fluid (CVF) was collected from eight
pregnant (mean gestation of 20 weeks) and eight non-pregnant women.
The CVF bacterial cell pellet was stained with IgG-Alexa Fluor. Opsonised
and non-opsonised bacteria were sorted using a MoFlow cell sorter.
Bacteria DNA was extracted from the sorted samples and composition
assessed using MiSeq based 16S rRNA gene surveying the V1-V2 region.
Cytokine, complement and immunoglobulin subtyping immunoassays
were performed on matched CVF supernatant.
RESULTS: All pregnant women delivered at term. The most abundant
immunoglobulin in CVF was IgG. Similar percentages of IgG opsonised
bacteria were observed in pregnant (46.2%) and non-pregnant women
(42.6%). IgG opsonisation of L. crispatus dominant communities in
pregnancy was variable, however G. vaginalis dominance correlated
with a high degree of IgG opsonisation, high inflammatory cytokine
expression and high IgA and IgM production. Complement activation in
the CVF varied according to vaginal microbiota composition, however
was generally suppressed in pregnancy. Similarly immunoglobulin
concentrations were consistently lower in pregnant CVF samples.
CONCLUSION: Fluorescence activated bacterial cell sorting and 16S
rRNA sequencing permits evaluation of vaginal microbiota and the
host immune response. High levels of G. vaginalis opsonisation may
reflect a heightened immune response in attempt to eliminate pathogenic
bacteria. The variable degree of opsonisation of different bacterial species
may reflect patient variability in adaptive immunity and individual
susceptibility to microbial associated risk of preterm delivery.

Scientific Abstracts

T-019
Fever in the Term Parturient and Epidural Analgesia with Absent
Placental Inflammation. William M Curtin*,1,2 Heather Florescue,2 Philip
J Katzman,2 Leon A Metlay,2 Avi Hameroff,1 Odessa Hamidi†,1 Serdar
H Ural.1 1Penn State College of Medicine, Hershey, PA, United States;
2
University of Rochester School of Medicine & Dentistry, Rochester, NY,
United States.
INTRODUCTION: Epidural analgesia and histologic chorioamnionitis
are independent risk factors for intrapartum fever at term. We sought to
determine which labor factors were associated with fever in women with
epidural analgesia in the absence of histologic chorioamnionitis.
METHODS: The study was a reanalysis of a prior retrospective cohort
study of 641 term parturients evaluating clinical signs for accuracy in the
diagnosis of histologic chorioamnionitis. We excluded subjects that did
not have epidural analgesia and subjects with histologic chorioamnionitis.
Fever was defined by maternal tympanic temperature of ≥ 38°C. Epidural
analgesia was administered with an initial test dose of 45 mg of lidocaine
with 15 mcg epinephrine. This was followed by an initial bolus of 100
mcg of fentanyl and 20 mg bupivacaine followed by a continuous infusion
with 8.75 mg of bupivacaine and 28 mcg of fentanyl per hour. Data
were analyzed by t-tests and chi-square tests as appropriate. Logistic
regression analysis was used to evaluate selected independent variables
in the prediction of fever. Significance was set at p < 0.05.
RESULTS: There were 32 (17.3%) subjects with fever and 153
without fever. Comparisons of the cohorts are given in the Table. A
logistic regression analysis was performed to ascertain the effects of
temperature upon admission, six or more vaginal exams, duration of
both ruptured membranes and labor, and parity ≥ 1 on the likelihood
of developing intrapartum fever. The logistic regression model was
statistically significant X2 (5)=14.28, p=0.0141. The model explained
12.3 % (Nagelkerke R2) of the variance in fever. Only duration of
ruptured membranes was significantly associated with the likelihood of
developing intrapartum fever. Odds ratio= 1.05, 95% confidence interval:
1.01, 1.10, p=.0246.
CONCLUSION: Increasing duration of ruptured membranes is associated
with an increased likelihood of developing intrapartum fever in the
term parturient with epidural analgesia even in the absence of placental
inflammation. It is possible that the presence of sufficient amniotic
fluid may blunt the rise in temperature in laboring women with epidural
analgesia. Further research is needed into the the factors involved in
epidural analgesia itself as well as other maternal factors that could
contribute to intrapartum fever.
Intrapartum Fever vs. No Fever Comparisons
Variable

Fever(n=32)

No
Fever(n=153)

Maternal age

28.0 ± 6.8

28.0 ± 6.5

Gestational age

38.9 ± 1.37

38.9 ± 1.31

Parity ≥ 1

12 (37.5%)

78 (51%)

0.58
(0.30,1.34)

0.165

Caucasian

26 (81.3%)

118 (77.1%)

1.29
(0.46,3.80)

0.609

Induction of labor

30 (62.5%)

88 (56.5%)

1.23
(0.53,2.90)

0.603

Six or more vaginal
exams

17 (53.1%)

62(40.5%)

1.66(0.73,3.83)

0.190

Internal fetal
monitoring

18 (56.3%)

82 (53.6%)

1.11
(0.049,2.57)

0.784

Temperature on
admission

36.9 ± 0.3

36.8 ± 0.4

0.161

Labor, duration

19.1 ± 18.7

12.9 ±11.1

0.081

ROM, duration

12.5 ± 11.7

7.5 ± 6.9

Amnioinfusion

4 (12.5%)

26 (17.0%)

Birthweight

3.41 ± 0.50

3.3 ± 0.59

Cesarean delivery

17 (53.1%)

49 (32.0%)

OR (95%CI)

P-value
0.99
0.846

0.026
0.69 (0.19,
2.34)

0.531
0.302

2.41 (1.04,
5.59)

0.023

Table of Contents for the Digital Edition of SRI Supplement to Reproductive Sciences - Volume 25 Number 1 - March 2018