Original Paper

Fetal Diagn Ther Received: July 15, 2016

Accepted after revision: April 21, 2017
DOI: 10.1159/000475926
Published online: May 24, 2017

Contribution of Amniotic Fluid along Gestation

to the Prediction of Perinatal Mortality in
Women with Early Preterm Premature Rupture of
Membranes
Teresa Cobo a, b Jordina Munrós a José Ríos c, d Janisse Ferreri a
Federico Migliorelli a Núria Baños a Eduard Gratacós a, b Montse Palacio a, b
a
BCNatal – Barcelona Center for Maternal-Fetal and Neonatal Medicine (Hospital Clínic and Hospital Sant Joan de
Déu), Fetal i+D Fetal Medicine Research Center, IDIBAPS, University of Barcelona, b Center for Biomedical Research
on Rare Diseases (CIBER-ER), c Biostatistics and Data Management Core Facility, IDIBAPS, Hospital Clinic, and
d
Biostatistics Unit, Faculty of Medicine, Universitat Autónoma de Barcelona, Barcelona, Spain

Keywords When evaluating the LVP at different time points of gesta-

Early preterm premature rupture of membranes · Largest
vertical pocket · Amniotic fluid · Ultrasound · Perinatal
mortality
Abstract
Introduction: To evaluate the largest vertical pocket (LVP) of
amniotic fluid as a time-dependent factor to predict perina-
tal mortality in women with early preterm premature rup-
ture of membranes (EPPROM). Material and Methods: Ob-
servational cohort study of singleton pregnancies with
EPPROM <24 weeks. Termination of pregnancy (TOP) was
considered if the LVP was <2 cm at 7 days. The maternal and
neonatal characteristics of ongoing pregnancies were re-
corded. Prediction of perinatal mortality was estimated
based on the influence of the LVP as a time-dependent factor
after adjustment for maternal age, prior invasive procedure,
and gestational age at EPPROM. Results: Of 104 women, 39
requested TOP. Neonatal survival to discharge was 40%, in-
creasing to 74% if pregnancies achieved 24 weeks. LVP at
admission <1 cm, latency to delivery, and gestational age at
delivery were independent predictors of perinatal mortality.
tion, the highest perinatal mortality risk was established at 2
weeks (odds ratio 14.67, p < 0.001) after membrane rupture,
being 5.75 (p = 0.05) the week after and 10.93 (p = 0.037) be-
yond 2 weeks of EPPROM. Discussion: When LVP measure-
ment, gestational age at EPPROM, maternal age, and prior
invasive procedure were considered, we found that the
worst prognosis related to perinatal mortality was at 2 weeks
after EPPROM. © 2017 S. Karger AG, Basel
Introduction
Early preterm premature rupture of membranes
(EPPROM) is an uncommon complication with an inci-
dence of less than 1% [1]. Classically, neonatal survival to
discharge was 32–63% [2–8]. However, it substantially
improves at viability [9, 10].
Due to the consequences of EPPROM for the fetus,
prenatal counseling has become a challenge for obstetri-
T.C. and J.M. contributed equally to this paper.
130.102.42.98 - 6/9/2017 1:31:11 AM
University of Queensland

© 2017 S. Karger AG, Basel Dr. Teresa Cobo

Sabino de Arana 1
ES–08028 Barcelona (Spain)
Downloaded by:

E-Mail karger@karger.com
E-Mail tcobo @ clinic.ub.es
www.karger.com/fdt
cians and an issue for parents. Despite improvements in
ultrasound (US) equipment, clinical factors are still con-
sidered the main predictors of perinatal mortality, with
severe and prolonged oligohydramnios being among the
strongest factors [1, 3, 7, 11–14].
Severe oligohydramnios [15–17] has been defined as a
condition in which the largest vertical pocket (LVP) of
amniotic fluid measures <1 cm. However, the definition
of prolonged oligohydramnios remains unclear [2, 4, 9,
10, 15, 16, 18]. A precise definition of prolonged oligohy-
dramnios would help obstetricians and parents in the de-
cision-making process regarding the perinatal mortality.
Data specifically focusing on the evolution of the amni-
otic fluid along gestation are few and contradictory. Thus,
Hadi et al. [19] evaluated the role of the mean of amni-
otic fluid volume throughout pregnancy showing a high-
er neonatal survival to discharge in women with an ade-
quate amniotic fluid volume until delivery. However,
Brumbaugh et al. [9] recently observed that neonatal sur-
vival to discharge in prolonged EPPROM was not related
to the evolution of the LVP along time. They defined pro-
longed EPPROM as latency to delivery of ≥7 days. Ac-
cording to Brumbaugh et al. [9], the main determinant of
neonatal survival to discharge was gestational age at de-
livery. Likewise, Hunter et al. [18] analyzed the amount
of amniotic fluid as a predictor of neonatal survival to
discharge and found it did not predict neonatal survival.
Therefore, taking into account the controversy regard-
ing the role of prolonged oligohydramnios as a risk factor
of perinatal mortality, the aim of the study was to evaluate
the influence of the LVP of amniotic fluid at different
time points of gestation as a predictor of perinatal mortal-
ity.
Material and Methods
Study Population
We conducted a single-center observational analysis of a co-
hort of women admitted to the Maternal-Fetal Medicine Depart-
ment at Hospital Clinic of Barcelona with a diagnosis of EPPROM
before 24 weeks of gestation from December 2000 to December
2013. Gestational age was calculated based on the crown-to-rump
length at the first-trimester US.
All women with a singleton pregnancy admitted with a diagno-
sis of EPPROM in which clinical chorioamnionitis or microbial
invasion of the amniotic cavity (MIAC) was previously excluded
were considered eligible for the study. Multiple pregnancies, struc-
tural/chromosomal anomalies, fetal hydrops, and women with a
non-confirmed single episode of hydrorrhea were excluded.
EPPROM was defined as the leakage of amniotic fluid on spec-
ulum examination, confirmed by alkaline pH in the absence of
vaginal infection and in cases of doubt, by detection of the insulin-
2 Fetal Diagn Ther
DOI: 10.1159/000475926
like growth factor-binding protein-1 (IGFBP-1) in the vaginal for-
nix using a one-step immunochromatographic dipstick test (Ac-
tim PROM test®, Medix Biomedica or Amnioquick®, Biosynex,
SA) or, exceptionally, the instillation of fluorescein (1 mL per 9 mL
of saline solution) in the amniotic cavity.
Clinical chorioamnionitis was defined according to Gibbs’ cri-
teria [20] as temperature ≥37.8 ° C and two or more of the follow-
ing criteria: uterine tenderness, malodorous vaginal discharge, fe-
tal tachycardia (>160 beats/min), maternal tachycardia (>100
beats/min), and leukocytosis >15,000/mm3. MIAC was defined as
a positive aerobic/anaerobic amniotic fluid culture for bacteria or
yeast (chocolate agar for aerobes, Schaedler agar for anaerobes and
thioglycollate broth), or the presence of Ureaplasma spp. and My-
coplasma hominis in mycoplasma culture (Mycoplasma IST 2, bio-
Mérieux).
The pregnancy outcomes reported were: termination of preg-
nancy (TOP), perinatal mortality (including second-trimester
miscarriage [<24 weeks], stillbirth [≥24 weeks], and neonatal
death), gestational age at delivery, latency from EPPROM to deliv-
ery, pulmonary hypoplasia, skeletal deformities, and clinical cho-
rioamnionitis.
Pulmonary hypoplasia was clinically defined by the onset of
severe respiratory insufficiency immediately after delivery requir-
ing high ventilatory pressures. Chest X-ray findings suggestive of
pulmonary hypoplasia included small lungs with a diaphragm el-
evation above the seventh rib, bell-shaped chest, and downward-
sloping ribs. At autopsy, criteria for the diagnosis of pulmonary
hypoplasia were a lung-to-body weight ratio ≤0.012 or abnormal
radial alveolar counts [16]. Skeletal deformities were defined by the
presence of arthrogryposis or limb malposition.
Composite neonatal morbidity included any of the following:
respiratory distress syndrome defined by the presence of two or
more of the following criteria: persistent oxygen requirement for
>24 h, administration of exogenous surfactant and radiologic evi-
dence of hyaline membrane disease such as bilateral interstitial
infiltrate, opacification of the lung fields, pathological air broncho-
gram, or decreased lung volume; intraventricular hemorrhage
grade III–IV diagnosed by cranial US; necrotizing enterocolitis de-
fined as the radiologic finding of either intramural gas or free intra-
abdominal gas; early-onset neonatal sepsis defined by clinical and
analytical criteria, and neonatal death.
Prediction of perinatal mortality was estimated based on the
influence of the LVP as a time-dependent factor after adjustment
for maternal age, prior invasive procedure, and gestational age at
EPPROM.
Clinical Management
At admission, a maternal blood analysis including C-reactive
protein (CRP) and white blood cell (WBC) count was carried out.
Aerobic, anaerobic, and fungi cultures in the cervical secretions
were also performed as well as fetal US assessment including the
transvaginal measurement of cervical length and the LVP of am-
niotic fluid. From 2005, transabdominal amniocentesis was intro-
duced into the clinical management of EPPROM to rule out MIAC
in women diagnosed with PPROM from 22 to 34 weeks.
A 5-day management in an inpatient regimen was undertaken
with strict bed rest during the first 2 days. Afterwards, personal
hygiene was allowed.
A course of 5-day intravenous ampicillin (1 g/6 h) and genta-
micin (80 mg/8 h) was systematically given to all women with
130.102.42.98 - 6/9/2017 1:31:11 AM
Cobo/Munrós/Ríos/Ferreri/Migliorelli/
University of Queensland
Baños/Gratacós/Palacio
Downloaded by:
membrane rupture [21, 22], as well as a single dose of oral azithro-
mycin (1 g) at admission. Azithromycin was introduced after 2012
as targeted prophylactic treatment of the most common microor-
ganism isolated in women with MIAC [23], Ureaplasma spp.,
based on its efficacy reported in animal models and its pharmaco-
kinetics in pregnant women [24–26].
Antibiotic treatment was withdrawn in case of negative cul-
tures. Maternal and fetal status was closely monitored for signs of
clinical chorioamnionitis or labor. WBC count and CRP analysis
were performed daily during the first 3 days of admission and
weekly thereafter. Fetal heart rate was auscultated daily during
hospital stay.
US assessment was repeated 1 week after admission. After
counseling, TOP was considered according to national legislation
before 23 weeks upon parental request in fetuses with severe and
prolonged oligohydramnios, defined in our institution when the
LVP <2 cm observed at admission remained at 7 days after mem-
brane rupture.
Women who requested expectant management were offered
outpatient follow-up with instructions to rest at home and with
weekly US examinations and CRP and WBC determinations until
delivery. They were instructed to come to the emergency room in
case of bleeding, temperature >37.8 ° C, or uterine contractions.
LVP differential was defined as the difference between the LVP
at the last US study and the LVP at admission.
Two intramuscular doses of betamethasone 12 mg were admin-
istered 24 h apart beyond 24 weeks when clinical changes suggest-
ed early delivery (e.g., shortened cervical length, uterine contrac-
tions, bleeding, elevation of CRP levels or WBC count). From 2011
onwards, magnesium sulfate was administered for fetal neuropro-
tection if labor started between 24 and 32 weeks. Our policy of
neonatal resuscitation considered an active management above 24
weeks or neonatal birth weight greater than 500 g.
Labor was only induced after 32 weeks if lung maturity was
documented (by amniotic fluid TDx-Fetal Lung Maturity-II test
[27]). In the remaining women, labor was induced after 34 weeks.
Indications for early delivery before 34 weeks were: clinical cho-
rioamnionitis or non-reassuring fetal test (biophysical profile
score ≤4, or atypical or late decelerations of fetal heart rate). Indi-
cations for Cesarean delivery were non-reassuring fetal status or
non-vertex presentation.
Statistical Analysis
Statistical analysis was performed using the SPSS 20 (IBM Inc.,
USA). Continuous variables were compared using a nonparamet-
ric Mann-Whitney U test presented as median with interquartile
range (IQR: 25th; 75th percentiles). Categorical variables were
compared using the Fisher exact test and presented as number
(percentage).
Univariate logistic regression models were used to evaluate in-
dependent clinical factors for predicting perinatal mortality.
In order to evaluate the median LVP (cm) at admission, at 1
week, at 2 weeks, and >2 weeks, a generalized estimated equation
(GEE) model was performed with a nonparametrical approach by
means of rank transformation for LVP (cm).
An approach to predict perinatal mortality was performed with
a GEE model using an AR(1) matrix in order to estimate the intra-
subject correlation. This longitudinal prediction estimates the risk
of end point (perinatal mortality) by mean of odds ratios (OR) and
their 95% confidence intervals (95% CI) for each independent fac-
Contribution of the Amniotic Fluid to
EPPROM Outcome
tor (LVP) and its follow-up (as number of weeks). Data was ad-
justed for maternal age, prior invasive procedure, and gestational
age at EPPROM. Additionally, ROC curves were performed by
weeks to estimate perinatal mortality using the final GEE model.
The area under the curve (AUC) and the 95% CI were calculated.
Differences were considered statistically significant with a two-
sided p value <0.05.
Ethics Committee
This study was approved by the Ethics Committee of Hospital
Clinic of Barcelona HCB/2015/0556.
Results
Over the study period, 154 women were admitted with
a diagnosis of EPPROM, 40 of whom carried multiple
pregnancies. Four women were admitted with symptoms
of clinical chorioamnionitis and 6 were diagnosed with
MIAC. Therefore, 104 women finally met the inclusion
criteria. Figure 1 shows the flow of events from admission
to neonatal discharge of the women finally included. The
median (IQR: 25th; 75th percentiles) gestational age at
EPPROM (weeks) of the entire study group was 18.5
(15.7; 21.3), with the LVP at admission (cm) being 1.4
(0.9–2.2).
Of the 104 women included at the beginning of the
study, 39 (37.5%) requested TOP because of persistence
of LVP <2 cm at 7 days after EPPROM. Comparison of
maternal characteristics correlated to TOP is shown in
the online supplementary appendix S1 (see www.
karger.com/doi/10.1159/000475926 for all online suppl.
material). Significant earlier gestational age at EPPROM
and a lower LVP at last US were observed in women who
requested TOP.
The baseline characteristics of the ongoing pregnan-
cies (65 cases) are summarized in Table 1. The median
(IQR: 25th; 75th percentiles) gestational age at admission
was 20.2 (17.1; 22.4). At admission, the LVP of amniotic
fluid was <1 cm in 20 women (30.8%), between 1–2 cm
in 24 (36.9%), and >2 cm in 21 (32.3%). Comparison of
perinatal outcomes correlated with gestational age at
EPPROM (before or after 20 weeks) is shown in the on-
line supplementary appendix S2.
Twenty-seven women had undergone a prior invasive
procedure: 8 transcervical chorionic villous sampling, 15
genetic amniocentesis, 1 woman with both CVS and am-
niocentesis, and 3 cervical cerclage. Online supplemen-
tary appendix S3 shows the comparison of perinatal data
according to prior invasive procedure.
130.102.42.98 - 6/9/2017 1:31:11 AM
Fetal Diagn Ther 3
University of Queensland
DOI: 10.1159/000475926
Downloaded by:
 n
120
104
100

80
65
60

35
40 33
26
20

0
Fig. 1. Flow of events from admission to EPPROM 1 week after 24 weeks Delivery Live newborn
neonatal discharge of the women finally in- <24 weeks EPPROM
cluded.

Table 1. Baseline characteristics of ongoing pregnancies Table 2. Neonatal outcomes of live newborns

Ongoing pregnancies Live newborns (n = 33)

(n = 65)
GA at delivery, weeks 29.2 (25.5; 32.5)
Maternal age, years 33 (28; 37) Birth weight, g 1,209.5 (760; 2,185)
Nulliparity 24 (37) Apgar <7 at 5 min 6 (18)
GA at EPPROM, weeks 20.2 (17.1; 22.4) Umbilical artery cord pH <7.1 2 (6)
Prior invasive procedure 27 (42) Composite neonatal morbidity 23 (70)
GA at invasive procedure, weeks 15.8 (13.2; 17) Admission to NICU, days 53 (26; 67)
LVP at admission, cm 1.5 (0.9; 2.5) Respiratory distress syndrome 15 (45)
Cervical length at admission, mm 34 (26; 37) Intraventricular hemorrhage
GA at delivery, weeks 24.2 (23.1; 29.2) grade III–IV 1 (3)
>24 weeks* 29.1 (25.1; 32.5) Necrotizing enterocolitis 1 (3)
Latency from EPPROM to delivery, days 42 (13; 59) Early-onset neonatal sepsis 9 (27)
>24 weeks* 56 (19; 98) Neonatal death 7 (21)
Second-trimester miscarriages 30 (46)
Stillbirth 2 (3) Continuous variables are presented as median (interquartile
Pulmonary hypoplasia 4 (6) range: 25th; 75th percentiles). Categorical variables are presented
Skeletal deformities 6 (9) as number (%). GA, gestational age; NICU, neonatal intensive care
Clinical chorioamnionitis 23 (35) unit.

Continuous variables are presented as median (interquartile

range: 25th; 75th percentiles). Categorical variables are presented
as number (%). GA, gestational age; EPPROM, early preterm
premature rupture of membranes; LVP, largest vertical pocket of
amniotic fluid. *GA at delivery or latency from EPPROM to
delivery only in pregnancies above 24 weeks.
Of these 65 ongoing pregnancies, only 35 delivered
≥24 weeks. Two stillbirths (at 25.1 and 25.3 weeks) and 7
neonatal deaths were reported. The overall survival rate
was 40% (26/65) increasing to 74% if pregnancies achieved
24 weeks of gestation (26/35). Preterm delivery before 34
weeks occurred in 31/35 (88.5%) women who delivered
above 24 weeks.
All 4 newborns with pulmonary hypoplasia died after
birth (2 at 23.2 weeks with a median LVP of amniotic
fluid <1 cm, 1 at 25.6 weeks with a median of LVP ob-
served >2 cm, and 1 at 29.2 weeks with a median of LVP
between 1–2 cm). Six skeletal deformities were observed
(2 due to arthrogryposis and 4 due to limb malposition).
Table 2 shows the neonatal outcomes of 33 live new-
borns. Five out of 7 neonatal deaths had an Apgar score
<7 at 5 min (1 also with umbilical artery cord pH <7.1).
The overall rate of clinical chorioamnionitis was 35%,
with no cases of maternal sepsis or death reported.
Clinical characteristics were evaluated by univariate
logistic regression as predictors of perinatal mortality
130.102.42.98 - 6/9/2017 1:31:11 AM

4 Fetal Diagn Ther Cobo/Munrós/Ríos/Ferreri/Migliorelli/

University of Queensland

DOI: 10.1159/000475926 Baños/Gratacós/Palacio

Downloaded by:
 Table 3. Clinical characteristics as predictors of perinatal mortality in ongoing pregnancies

Perinatal mortality Neonatal survival to OR (95% CI) p

(n = 39) discharge (n = 26)

Maternal age, years 35 (29; 38) 32 (28; 36) 1 (0.9–1.2) 0.828

Prior invasive procedure 13 (33) 14 (54) 0.4 (0.2–1.2) 0.103
GA at EPPROM, weeks 20.6 (18.3; 22.3) 19.3 (16; 23) 1.1 (0.9–1.2) 0.492
LVP at admission, cm 1.1 (0.8; 2) 2 (1.3; 3) 0.6 (0.4–0.9) 0.026
LVP at admission <1 cm 18 (46) 4 (15) 4.7 (1.2–17.8) 0.024
LVP at admission 1–2 cm 11 (28) 11 (42) 1.6 (0.5–5.3) 0.438
LVP at admission ≥2 cm 10 (26) 11 (42) 1
LVP at last US assessment
before delivery, cm 1 (0.5; 2.3) 1.85 (1.1; 3) 0.6 (0.4–1) 0.035
LVP differential* 0 (–0.9 to 1.1) 0 (–1.3 to 0.3) 1.6 (0.2–11.4) 0.661
Latency from EPPROM to
delivery, days 13 (7; 21.7) 77 (35; 130.9) 0.95 (0.93–0.98) <0.001
GA at delivery, weeks 23.2 (21.1; 24) 30 (26.3; 36.4) 0.5 (0.3–0.7) <0.001

Continuous variables are described as median (interquartile range: 25th; 75th percentiles) except for LVP
differential (*), where median and range (min. to max.) are described. Variables were compared using Mann-
Whitney U test. Categorical variables were compared using Fisher exact test and presented as number (%).
Univariate logistic regression was performed to assess the relationship between clinical variables and the
occurrence of perinatal mortality. OR, odds ratio; CI, confidence interval; GA, gestational age; EPPROM, early
preterm premature rupture of membranes; LVP, largest vertical pocket of amniotic fluid; US, ultrasound.

Table 4. Largest vertical pocket (LVP) at weekly follow-up from admission to delivery (n = 65)

Admission Follow-up
at 1 week at 2 weeks >2 weeks

Perinatal mortality
n 39 33 18 3
LVP, cm 1.1 (0.8; 2) 1.5 (0.9; 3) 1.6 (0.9; 2.5) 0.5 (0.13; 3)
p value (from admission) 0.073 0.527 0.508
Neonatal survival to discharge
n 26 26 26 19
LVP, cm 2 (1.3; 3) 2 (1.5; 3.1) 2.1 (1; 3.8) 2 (1.1; 3)
p value (from admission) 0.332 0.716 0.760

LVP presented as median (interquartile range: 25th; 75th percentiles); p > 0.05.

(Table 3). Only LVP of amniotic fluid at admission <1 cm,
LVP at the last US exploration previous to delivery, la-
tency from EPPROM to delivery, and gestational age at
delivery were found to be significant predictors of perina-
tal mortality, while gestational age at EPPROM and LVP
differential were not.
Table 4 shows the evolution of the median LVP during
US follow-up. The median of LVP was not significantly
modified at different time points of gestation either in the
“perinatal mortality” group or in the “survival to dis-
charge” group.
When the LVP of amniotic fluid observed at different
time points was <1 cm, the risk of perinatal mortality
showed an OR of 4.64 (95% CI 1.77–12.14, p = 0.002).
However, we did not observe a higher risk of perinatal
mortality when the median of the LVP was between 1 and
2 cm along weekly US evaluations (OR 1.87; 95% CI 0.74–
4.76, p = 0.189).
130.102.42.98 - 6/9/2017 1:31:11 AM

Contribution of the Amniotic Fluid to Fetal Diagn Ther 5

University of Queensland

EPPROM Outcome DOI: 10.1159/000475926

Downloaded by:
 Table 5. Association between largest vertical pocket (LVP) at weekly ultrasound and perinatal mortality

Admission Follow-up
at 1 week at 2 weeks >2 weeks

Perinatal mortality
LVP
<1 cm 16 (41) 12 (36.4) 5 (27.8) 2 (66.7)
1–2 cm 14 (35.9) 7 (21.2) 8 (44.4) 0
>2 cm 9 (23.1) 14 (42.4) 5 (27.8) 1 (33.3)
Neonatal survival to discharge
LVP
<1 cm 4 (15.4) 2 (7.7) 5 (19.2) 3 (15.8)
1–2 cm 10 (38.5) 12 (46.2) 6 (23.1) 8 (42.1)
>2 cm 12 (46.2) 12 (46.2) 15 (57.7) 8 (42.1)

OR (95% CI)* 1 5.75 (1.68–19.74) 14.67 (3.54–60.82) 10.93 (1.16–103.08)

p value 0.005 <0.001
AUC (95% CI) 0.58 (0.36–0. 79) 0.8 (0.68–0.92) 0.84 (0.72–0.95) 0.79 (0.56–1)
p value 0.541 0.001 <0.001 0.114

Data presented as number (%). * Odds ratio (OR) adjusted by maternal age, prior invasive procedure, and
gestational age at early preterm premature rupture of membranes. AUC, area under the curve; CI, confidence
interval.

Finally, the prediction of perinatal mortality was based
on the measurement of the median of the LVP as a time-
dependent factor (at 1 week, at 2 weeks, and more than 2
weeks after EPPROM) (Table 5). Data were adjusted for
maternal age, prior invasive procedure, and gestational
age at EPPROM and found that the worst prognosis re-
lated to perinatal mortality was at 2 weeks after mem-
brane rupture (OR 14.67; 95% CI 3.54–60.82, p < 0.001)
with an AUC and a 95% CI to predict perinatal mortality
of 0.84 and 0.72–0.95. However, at 1 week after EPPROM,
the AUC was also high and similar to that observed at
week 2 with an AUC of 0.8 (95% CI 0.68–0.92) (Table 5).
The estimated increased risk of perinatal mortality given
beyond 2 weeks did not significantly differ when com-
pared with the prognosis at 2 weeks.
Discussion
Our data highlight the importance of the LVP of am-
niotic fluid as a time-dependent factor that may help to
predict perinatal mortality. Our results suggest that al-
though at 1 week after EPPROM the risk of perinatal
mortality in women with severe oligohydramnios is high,
the highest risk is observed 2 weeks after membrane rup-
ture.
Similar to data previously reported by our group [28]
and others [11, 29], our neonatal survival rate was 40%,
which substantially rose to 74% when pregnancies
achieved 24 weeks of gestation, probably due to manage-
ment with antenatal corticosteroids and postnatal strate-
gies after viability. Regarding other clinical outcomes re-
lated to EPPROM, our rate of pulmonary hypoplasia of
6% is slightly lower than previously reported [1, 4, 11].
Skeletal deformities were observed in 9% of women
(6/65), similar to other studies reporting a rate ranging
from 1.5 to 38% [11]. With regard to maternal complica-
tions, the overall rate of clinical chorioamnionitis was
35%, very similar to previously published data [11].
Although the role of amniotic fluid as a predictor of
mortality has recently been questioned [9, 18], our data,
including a larger number of women, and in line with
previous reports [1, 4, 12, 30, 31], suggest that the pres-
ence of a lower LVP at admission is associated with a sig-
nificantly higher rate of perinatal mortality.
Similar to Brumbaugh et al. [9], we also found an ear-
lier gestational age at delivery in women with a higher
perinatal mortality, emphasizing the importance of ges-
tational age at delivery as a strong predictor of mortality
regardless of other clinical factors.
On the other hand, contrary to the findings of van der
Heyden et al. [32], we and Brumbaugh et al. [9] did not
130.102.42.98 - 6/9/2017 1:31:11 AM

6 Fetal Diagn Ther Cobo/Munrós/Ríos/Ferreri/Migliorelli/

University of Queensland

DOI: 10.1159/000475926 Baños/Gratacós/Palacio

Downloaded by:
find gestational age at EPPROM to influence perinatal
mortality. Our inclusion criteria were much stricter than
theirs, which included women with PPROM until 27
weeks. Indeed, our results suggest that the main clinical
factors related to perinatal mortality are the LVP at ad-
mission and before delivery, the latency from EPPROM
to delivery, and the gestational age at delivery. Unfortu-
nately, in the antenatal counseling of these women, nei-
ther gestational age at delivery nor latency to delivery can
be taken into consideration.
Contrary to the previously reported 25% of re-accu-
mulation of amniotic fluid in women with EPPROM [19],
we did not observe significant changes in the measure-
ment of the LVP when evaluated at different time points
of gestation. Indeed, the median LVP remained almost
constant along US evaluation.
Since the LVP at admission and before delivery are sig-
nificantly associated with a higher rate of perinatal mor-
tality, we considered it crucial to include the factor of time
in the evaluation of the amniotic fluid. In order to define
the term “prolonged” in women with severe oligohy-
dramnios, we evaluated the influence of LVP on the pre-
diction of perinatal mortality during the weekly follow-
up before delivery after adjustment for variables such as
maternal age, prior invasive procedure, and gestational
age at EPPROM. Accordingly, although the risk of mor-
tality at 1 week after membrane rupture was high, the
highest risk of perinatal mortality was observed when se-
vere oligohydramnios was found at 2 weeks after mem-
brane rupture.
The main strength of the study was the prospective
follow-up of ongoing pregnancies allowing US informa-
tion in relation to the LVP evolution until delivery.
Nevertheless, our study also has some limitations. It
was an observational study (level 2b) performed in a sin-
gle institution, thereby limiting the sample size. None-
theless, EPPROM is a rare complication and thus, the
number of women reported in other studies is similar to
ours [5–10]. Another limitation is that almost one-third
of the women requested TOP. Additionally, our results
were obtained following a clinical protocol that included
amniocentesis to rule out MIAC above 22 weeks of gesta-
tion. Thus, on excluding women with MIAC or TOP, the
rates of severe complications such as perinatal mortality
or pulmonary hypoplasia were probably underestimated.
Moreover, the potential contribution of the LVP at dif-
ferent time points of gestation in these women who re-
quested a TOP is unknown. In addition, neonatal out-
come might be influenced by the introduction of magne-
sium sulfate, a different antibiotic regimen over time, or
Contribution of the Amniotic Fluid to
EPPROM Outcome
the assessment of lung maturity with amniocentesis. An-
other limitation was the small sample size at 2 weeks
when LVP was evaluated at different time points of ges-
tation. Finally, the 42% of iatrogenic EPPROM in our
population may explain the lack of association between
gestational age at EPPROM and perinatal mortality,
making our findings not generalizable to other popula-
tions where the number of iatrogenic EPPROM is much
lower.
In conclusion, we developed a prediction model that
estimates the risk of perinatal mortality for the mea-
surement of LVP and its follow-up (at 1 week, at 2
weeks, more than 2 weeks) being adjusted for other
variables such as maternal age, prior invasive proce-
dure, and gestational age at EPPROM. According to our
prediction model, the highest risk of perinatal mortality
was found at 2 weeks after membrane rupture, although
the prediction was also high at the first week. Thus, in
institutions where the management of EPPROM is sim-
ilar to ours, prenatal counseling related to perinatal
mortality prediction may be done at the first or second
week of EPPROM. There is no advantage to waiting
more than 2 weeks to give a more exact prognosis in
these women.
Further studies including a multicenter design with a
larger sample size are needed to validate these findings
and to establish whether the best time point to predict
perinatal mortality is at 1 or at 2 weeks from EPPROM.
Acknowledgements
The authors would like to acknowledge the Antenatal Clinic at
the Maternal and Fetal Medicine Department of Hospital Clinic,
Barcelona, Spain. Funding for publication was supported by Span-
ish grants from the Plan Estatal de Investigación Científica y Téc-
nica y de Innovación (I+D+i) 2013–2016 (PI10/01308, PI15/00344),
funded by Instituto de Salud Carlos III (ISCIII) – Subdirección
General de Evaluación and by the Fondo Europeo de Desarrollo
Regional (FEDER).
Disclosure Statement
The authors declare that there are no conflicts of interest.
130.102.42.98 - 6/9/2017 1:31:11 AM
Fetal Diagn Ther 7
University of Queensland
DOI: 10.1159/000475926
Downloaded by:
 References
1 Mercer BM: Preterm premature rupture of
the membranes. Obstet Gynecol 2003; 101:
178–193.
2 Rib DM, Sherer DM, Woods JR Jr: Maternal
and neonatal outcome associated with pro-
longed premature rupture of membranes be-
low 26 weeks’ gestation. Am J Perinatol 1993;
10:369–373.
3 Kilbride HW, Yeast J, Thibeault DW: Defin-
ing limits of survival: lethal pulmonary hypo-
plasia after midtrimester premature rupture
of membranes. Am J Obstet Gynecol 1996;
175:675–681.
4 Winn HN, Chen M, Amon E, Leet TL, Shum-
way JB, Mostello D: Neonatal pulmonary hy-
poplasia and perinatal mortality in patients
with midtrimester rupture of amniotic mem-
branes – a critical analysis. Am J Obstet Gy-
necol 2000;182:1638–1644.
5 Xiao ZH, Andre P, Lacaze-Masmonteil T, Au-
dibert F, Zupan V, Dehan M: Outcome of pre-
mature infants delivered after prolonged pre-
mature rupture of membranes before 25
weeks of gestation. Eur J Obstet Gynecol Re-
prod Biol 2000;90:67–71.
6 Taylor J, Garite TJ: Premature rupture of
membranes before fetal viability. Obstet Gy-
necol 1984;64:615–620.
7 Major CA, Kitzmiller JL: Perinatal survival
with expectant management of midtrimester
rupture of membranes. Am J Obstet Gynecol
1990;163:838–844.
8 Dinsmoor MJ, Bachman R, Haney EI, Gold-
stein M, Mackendrick W: Outcomes after ex-
pectant management of extremely preterm
premature rupture of the membranes. Am J
Obstet Gynecol 2004;190:183–187.
9 Brumbaugh JE, Colaizy TT, Nuangchamnong
N, O’Brien EA, Fleener DK, Rijhsinghani A,
Klein JM: Neonatal survival after prolonged
preterm premature rupture of membranes
before 24 weeks of gestation. Obstet Gynecol
2014;124:992–998.
10 Williams O, Hutchings G, Debieve F, De-
bauche C: Contemporary neonatal outcome
following rupture of membranes prior to 25
weeks with prolonged oligohydramnios. Ear-
ly Hum Dev 2009;85:273–277.
11 Waters TP, Mercer BM: The management of
preterm premature rupture of the membranes
near the limit of fetal viability. Am J Obstet
Gynecol 2009;201:230–240.
12 Vergani P, Ghidini A, Locatelli A, Cavallone
M, Ciarla I, Cappellini A, Lapinski RH: Risk
factors for pulmonary hypoplasia in second-
trimester premature rupture of membranes.
Am J Obstet Gynecol 1994;170:1359–1364.
8 Fetal Diagn Ther
DOI: 10.1159/000475926
13 Lao TT, Cheung VY: Expectant management
of preterm prelabour rupture of membranes
– the significance of oligohydramnios at pre-
sentation. Eur J Obstet Gynecol Reprod Biol
1993;48:87–91.
14 Lee JY, Ahn TG, Jun JK: Short-term and long-
term postnatal outcomes of expectant man-
agement after previable preterm premature
rupture of membranes with and without per-
sistent oligohydramnios. Obstet Gynecol
2015;126:947–953.
15 Bastide A, Manning F, Harman C, Lange I,
Morrison I: Ultrasound evaluation of amni-
otic fluid: outcome of pregnancies with severe
oligohydramnios. Am J Obstet Gynecol 1986;
154:895–900.
16 Barss VA, Benacerraf BR, Frigoletto FD Jr:
Second trimester oligohydramnios, a predic-
tor of poor fetal outcome. Obstet Gynecol
1984;64:608–610.
17 Shipp TD, Bromley B, Pauker S, Frigoletto FD
Jr, Benacerraf BR: Outcome of singleton preg-
nancies with severe oligohydramnios in the
second and third trimesters. Ultrasound Ob-
stet Gynecol 1996;7:108–113.
18 Hunter TJ, Byrnes MJ, Nathan E, Gill A, Pen-
nell CE: Factors influencing survival in pre-
viable preterm premature rupture of mem-
branes. J Matern Fetal Neonatal Med 2012;25:
1755–1761.
19 Hadi HA, Hodson CA, Strickland D: Prema-
ture rupture of the membranes between 20
and 25 weeks’ gestation: role of amniotic fluid
volume in perinatal outcome. Am J Obstet
Gynecol 1994;170:1139–1144.
20 Gibbs RS, Blanco JD, St Clair PJ, Castaneda
YS: Quantitative bacteriology of amniotic flu-
id from women with clinical intraamniotic in-
fection at term. J Infect Dis 1982;145:1–8.
21 Hopkins L, Smaill F: Antibiotic regimens for
management of intraamniotic infection. Co-
chrane Database Syst Rev 2002;(3):CD003254.
22 Cararach V, Botet F, Sentis J, Almirall R, Per-
ez-Picanol E: Administration of antibiotics to
patients with rupture of membranes at term:
a prospective, randomized, multicentric
study. Collaborative Group on PROM. Acta
Obstet Gynecol Scand 1998;77:298–302.
23 Cobo T, Palacio M, Martinez-Terron M, Na-
varro-Sastre A, Bosch J, Filella X, Gratacos E:
Clinical and inflammatory markers in amni-
otic fluid as predictors of adverse outcomes in
preterm premature rupture of membranes.
Am J Obstet Gynecol 2011;205:126.e1–e8.
24 Grigsby PL, Novy MJ, Sadowsky DW, Mor-
gan TK, Long M, Acosta E, Duffy LB, Waites
KB: Maternal azithromycin therapy for Urea-
plasma intraamniotic infection delays pre-
term delivery and reduces fetal lung injury in
a primate model. Am J Obstet Gynecol 2012;
207:475.e1–e475.e14.
25 Kemp MW, Miura Y, Payne MS, Watts R,
Megharaj S, Jobe AH, Kallapur SG, Saito M,
Spiller OB, Keelan JA, et al: Repeated mater-
nal intramuscular or intraamniotic erythro-
mycin incompletely resolves intrauterine
Ureaplasma parvum infection in a sheep
model of pregnancy. Am J Obstet Gynecol
2014;211:134.e1–e9.
26 Ramsey PS, Vaules MB, Vasdev GM, An-
drews WW, Ramin KD: Maternal and trans-
placental pharmacokinetics of azithromycin.
Am J Obstet Gynecol 2003;188:714–718.
27 Bennasar M, Figueras F, Palacio M, Bellart J,
Casals E, Figueras J, Coll O, Gratacos E: Ges-
tational age-specific cutoff levels of TDx-FLM
II for the prediction of neonatal respiratory
distress syndrome. Fetal Diagn Ther 2009;25:
392–396.
28 Palacio M, Cobo T, Figueras F, Gomez O, Coll
O, Cararach V, Gratacos E: Previable rupture
of membranes: effect of amniotic fluid on
pregnancy outcome. Eur J Obstet Gynecol Re-
prod Biol 2008;138:158–163.
29 Mercer BM, Goldenberg RL, Meis PJ, Moawad
AH, Shellhaas C, Das A, Menard MK, Caritis
SN, Thurnau GR, Dombrowski MP, et al: The
Preterm Prediction Study: prediction of pre-
term premature rupture of membranes
through clinical findings and ancillary testing.
The National Institute of Child Health and
Human Development Maternal-Fetal Medi-
cine Units Network. Am J Obstet Gynecol
2000;183:738–745.
30 Muris C, Girard B, Creveuil C, Durin L, Herli-
coviez M, Dreyfus M: Management of prema-
ture rupture of membranes before 25 weeks.
Eur J Obstet Gynecol Reprod Biol 2007; 131:
163–168.
31 Storness-Bliss C, Metcalfe A, Simrose R, Wil-
son RD, Cooper SL: Correlation of residual
amniotic fluid and perinatal outcomes in
periviable preterm premature rupture of
membranes. J Obstet Gynaecol Can 2012; 34:
154–158.
32 van der Heyden JL, van der Ham DP, van
Kuijk S, Notten KJ, Janssen T, Nijhuis JG,
Willekes C, Porath M, van der Post JA, Hal-
bertsma F, et al: Outcome of pregnancies with
preterm prelabor rupture of membranes be-
fore 27 weeks’ gestation: a retrospective co-
hort study. Eur J Obstet Gynecol Reprod Biol
2013;170:125–130.
130.102.42.98 - 6/9/2017 1:31:11 AM
Cobo/Munrós/Ríos/Ferreri/Migliorelli/
University of Queensland
Baños/Gratacós/Palacio
Downloaded by: