International Journal of Gynecology and Obstetrics 132 (2016) 329331

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

CLINICAL ARTICLE

Carcinoembryonic antigen as a biomarker for meconium-stained

amniotic uid
Amir Mor a,, Reshef Tal a, Mohamad Irani a, Sandra McCalla a, Shoshana Haberman a,
Deepika Garg a, Birgitta Wajntraub b
a
b

Department of Obstetrics and Gynecology, Maimonides Medical Center, Brooklyn, NY, USA
Department of Obstetrics and Gynecology, Laniado Medical Center, Netanya, Israel

a r t i c l e

i n f o

Article history:
Received 12 March 2015
Received in revised form 10 July 2015
Accepted 9 November 2015
Keywords:
Carcinoembryonic antigen
Cutoff
Detection
Meconium-stained amniotic uid
Rupture of membranes

a b s t r a c t
Objective: To assess whether elevated carcinoembryonic antigen (CEA) concentration in amniotic uid can
indicate meconium-stained amniotic uid (MSAF). Methods: In a prospective cohort study, women with a
term singleton pregnancy who were in labor but had intact membranes were recruited at a center in Israel
over a 5-month period in 2013. Only women who subsequently underwent articial rupture of membranes
following a clear medical indication were included. Samples of amniotic uid, urine, and serum were collected.
Amniotic uid was examined by sight and classied as clear, MSAF, or undetermined. CEA concentration in the
samples was measured. Results: Among 81 participants, 45 had clear amniotic uid, 28 had MSAF, and
eight had undetermined amniotic uid. Mean CEA concentration was more than 10 times higher in MSAF
(2658 g/L, standard error 250) than in clear amniotic uid (238 g/L, standard error 29; P b 0.001). Receiver operating characteristic curve analysis demonstrated a sensitivity of 96% and a specicity of 100% for distinguishing
MSAF from clear amniotic uid at a CEA cutoff of 799.2 g/L. CEA concentrations in urine and serum were all
within the normal range (5 g/L), irrespective of amniotic uid status. Conclusion: High CEA concentrations in
amniotic uid can assist in the diagnosis of MSAF. These ndings could provide the basis for a bedside test to
detect MSAF following rupture of membranes.
2015 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction
The prenatal detection of meconium in the amniotic uid can alert a
caregiver or patient to intrauterine fetal distress. To date, a leakage of
meconium-stained amniotic uid (MSAF) has simply been detected by
sight by experienced medical personnel. However, the conrmation of
MSAF leakage can be challenging even for experienced medical personnel when a woman arrives at hospital after the rupture of membranes
(ROM) at home. The ability to conrm leakage of MSAF could be helpful
in some clinical settings.
A simple method to differentiate amniotic uid from other
body uids absorbed in a pad by measuring the concentration of
-fetoprotein (AFP) has been demonstrated previously [1]. The AFP
concentration is signicantly higher in amniotic uid than it is in
semen, urine, or normal vaginal discharge [1]. Quantication of the
AFP level in a pad that has absorbed uid of unknown origin could assist
in the retrospective conrmation of ROM.

Corresponding author at: Department of Obstetrics and Gynecology, Maimonides

Medical Center, 4802 10th Ave, Brooklyn, NY 11219, USA. Tel.: + 1 917 741 8299,
+ 1 718 283 6078; fax: + 1 718 283 7167.
E-mail address: morami12@gmail.com (A. Mor).

Another potential method is the measurement of carcinoembryonic

antigen (CEA). CEA is a normal component of meconium, but is also
found in clear amniotic uid [2]. Sarandakou et al. [3] and Lolis et al.
[4] have shown that the CEA level is higher in MSAF than in clear amniotic uid. However, they did not characterize the distribution of CEA
concentrations in amniotic uid or determine whether CEA could
serve as a biomarker for MSAF. We hypothesized that the CEA concentration will increase signicantly when meconium is secreted into the
amniotic uid. The goal of the current study was to examine whether
measurement of the CEA concentration in amniotic uid can help to
reliably differentiate leakage of MSAF from leakage of clear amniotic
uid following ROM.
2. Materials and methods
A prospective cohort study was conducted between January 28 and
June 16, 2013. Women with a term singleton pregnancy, intact membranes, no known medical problems, and a normal prenatal course
were recruited in the Labor and Delivery Unit of Laniado Medical Center,
Netanya, Israel. Exclusion criteria were a history of malignant or benign
gastrointestinal or urogenital tumors that could affect CEA levels in
serum, urine, and/or amniotic uid. Written informed consent was obtained from all participants. The study protocol was designed in

http://dx.doi.org/10.1016/j.ijgo.2015.07.032
0020-7292/ 2015 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.

330

A. Mor et al. / International Journal of Gynecology and Obstetrics 132 (2016) 329331

accordance with the Israel Public Health Regulations (Clinical Trials on

Human Subjects) and approved by the Institutional Review Board of
Laniado Medical Center.
Among recruited women, only those who underwent articial ROM
following a clear medical indication were included. Samples of amniotic
uid for each of these women were collected in a cup. Samples of serum
and urine were also collected. Serum samples were collected by venipuncture during active labor. Urine samples were collected through a
urethral catheter that was placed during the labor process.
The amniotic uid samples were independently examined by three
medical professionals (physicians and/or midwives) and were classied
as clear or MSAF on the basis of a consensus between the three independent examiners. When no consensus was reached, the sample was
classied as undetermined. Because there is no objective laboratory
or imaging test for MSAF, clinical evaluation by skilled medical personnel served as the denitive classication method. Quantitative measurement of the CEA concentration in the amniotic uid, serum, and urine
samples was performed using the AxSYM (Abbott Laboratories, Abbott
Park, IL, USA) chemiluminometric immunoassay.
The statistical analyses were performed using MedCalc version 11
(MedCalc Software, Ostend, Belgium). On the basis of the assumption
that the data for age, pregnancy duration, birth weight, and CEA concentration met the conditions of a normal distribution, the t test was used
for comparisons of these data. The 2 or Fisher exact tests were used
for the comparison of proportions. All signicance tests were twotailed. P b 0.05 was considered statistically signicant. Receiver operating characteristic (ROC) curves were plotted to assess whether CEA can
serve as a biomarker for MSAF. Sensitivity and specicity values were
calculated for the optimal CEA cutoff concentration determined by
ROC curve analysis. The undetermined amniotic uid samples were
not included in the sensitivity and specicity analyses.
3. Results
A total of 81 amniotic uid samples from 81 participants were sent
for CEA quantication. Following classication by the three independent examiners, 45 samples were deemed clear amniotic uid and 28
were MSAF. Bloody secretions were visible in 16 (20%) of the samples.
Because of a considerable amount of blood that might have masked
the typical greenish color of meconium, eight samples could not be
classied as clear amniotic uid or MSAF (no consensus was reached between the three examiners); therefore, these samples were classied as
undetermined amniotic uid.
The mean maternal age, birth weight, sex of the newborn, and mode
of delivery did not differ signicantly between the clear amniotic uid
and MSAF groups (Table 1). The mean pregnancy duration was higher

in the MSAF group, but the difference between the groups in the present
study was not statistically signicant (P = 0.057) (Table 1). The proportion of neonates with an Apgar score below 7 at 1 minute after delivery
was higher in the MSAF group than in the clear amniotic uid group
(P = 0.04) (Table 1). At 5 minutes after delivery, these neonates had
recovered and no difference was noted between the groups.
CEA concentrations in the clear amniotic uid, MSAF, and undetermined amniotic uid groups are shown in Fig. 1. The mean CEA concentration was 2658 g/L (standard error of the mean [SEM] 250; range
7016634) in the MSAF group and 238 g/L (SEM 29; range 18799)
in the clear amniotic uid group (P b 0.001).
Analysis of the ROC curve revealed that the CEA concentration in
amniotic uid was highly predictive of the presence of MSAF (area
under the curve 0.998; P b 0.001) (Fig. 2). The optimum CEA cutoff
level for the prediction of MSAF was 799.2 g/L, giving a sensitivity of
96% and a specicity of 100%.
The measured CEA concentrations were within the normal range
(5 g/L) in all tested serum (mean 1.0 g/L, SEM 0.6) and urine samples
(mean 0.8 g/L, SEM 0.6).
4. Discussion
The present ndings showed that the levels of CEA were signicantly
higher in the MSAF group than in the clear amniotic uid group. The
distributions of CEA concentrations in these two groups had minimal
overlap, and ROC curve analysis demonstrated that a high CEA concentration in amniotic uid can distinguish MSAF from clear amniotic uid
with high sensitivity and specicity. A CEA cutoff of 799.2 g/L had a
sensitivity of 96% and a specicity of 100% for the detection of MSAF.
Eight samples were classied as undetermined amniotic uid because of a lack of consensus between the three medical professionals.
The CEA levels of these samples were both above and below the proposed cutoff level. Thus, the three undetermined amniotic uid samples
containing a CEA concentration of 799.2 g/L or more could potentiality
indicate undiagnosed MSAF. However, the other ve samples with a
CEA concentration of less than 799.2 g/L could potentially have been
misdiagnosed as MSAF, leading to unnecessary interventions and the
use of additional healthcare resources.
In the present study, the CEA levels in serum and urine did not
rise above the normal levels ( 5 g/L). Even in urine samples from
women with MSAF, the CEA concentrations remained at normal

Table 1
Clinical characteristics and labor outcomes of patients with clear amniotic uid versus
MSAF.a
Variable

Clear amniotic uid

(n = 45)

MSAF
(n = 28)

P value

Maternal age, y
Pregnancy duration, wk
Parity
Birth weight, g
Sex of the newborn
Male
Female
Mode of delivery
Spontaneous
Cesarean delivery
Vacuum extraction
Apgar score b7 at 1 min
Apgar score b7 at 5 min

27.9 5.8
39.2 1.2
1.7 (09)
3357.2 513.6

28.1 6.1
39.9 1.9
1.8 (07)
3481.1 592.7

0.889
0.057
N/A
0.348
0.354

21 (47)
24 (53)

17 (61)
11 (39)

35 (78)
6 (13)
4 (9)
3 (7)
0

18 (64)
9 (32)
1 (4)
7 (25)
2 (7)

0.127

0.04
0.283

Abbreviations: MSAF, meconium-stained amniotic uid; N/A, not applicable.

a
Values are given as mean SD, mean (range), or number (percentage), unless indicated otherwise.

Fig. 1. CEA concentration in clear amniotic uid, MSAF, and undetermined samples.
Abbreviations: CEA, carcinoembryonic antigen; MSAF, meconium-stained amniotic uid.

A. Mor et al. / International Journal of Gynecology and Obstetrics 132 (2016) 329331

Fig. 2. Receiver operating characteristic curve for different carcinoembryonic antigen concentrations in amniotic uid. Area under the curve, 0.998; P b 0.001. The optimum cutoff
level was 799.2 g/L, with a sensitivity of 96% and a specicity of 100% for detecting meconium-stained amniotic uid. The blue dotted lines show 95% condence intervals.

levels, indicating that CEA does not leak from the amniotic uid to other
body uid compartments.
The sole presence of meconium in amniotic uid is not sufcient to
dictate clinical decisions during labor management. Although there is
a consensus that a pregnant woman with MSAF should be closely monitored for fetal well-being and that a pediatrician should be present during delivery [5], there is no evidence supporting obstetric intervention
to prevent meconium aspiration and its complications [6,7]. In agreement with previous ndings [8], the proportion of neonates with an
Apgar score below 7 at 1 minute was higher in the MSAF group in the
present study, supporting the recommendation for the presence of a pediatrician/neonatologist in the delivery room when MSAF is detected.
Moreover, awareness of the presence of MSAF might be invaluable in
the setting of category II fetal heart rate tracings. Additionally, the conrmation of ROM with MSAF can assist in selecting the optimal approach of labor management. For example, when MSAF is detected
before a planned delivery at home or in a birthing center, a transfer to
a hospital should be considered. It is also important to recognize that
this area is still under investigation and that early and more accurate detection of MSAF might ultimately prove to be of clinical importance.
There are a few limitations to the present study that should be
mentioned. One should bear in mind that a negative result (CEA concentration b 799.2 g/L) does not exclude the presence of meconium:
ROM could initially present as a leakage of clear amniotic uid.
However, during the course of active labor, the fetus may pass meconium that will remain in utero because ROM has already occurred and the
remaining amniotic uid surrounding the fetus is not sufcient to wash
out the viscous meconium. This is an inherent limitation of the present

331

method because uid leaking out of the introitus was collected rather
than actively sampling of the uid surrounding the fetus. Additionally,
it is important to note that various commercial immunoassays are available for CEA quantication in different body uids. Analysis of the same
specimen using different assays could provide different concentration
results, and therefore cutoff levels should be determined accordingly.
Finally, besides the Apgar score, no neonatal outcomes and complications were assessed. Future longitudinal studies are warranted to evaluate the correlation between the CEA concentration in amniotic uid and
neonatal outcomes.
Reliable tests based on the quantication of biomarker(s) that detect
specic condition(s) are of great importance. For instance, Barnhart [9]
commented on the need for a reliable biomarker for ectopic pregnancy.
The retrospective diagnosis of ROM by detection of amniotic uid in
sampling pads worn by pregnant women has been previously demonstrated [1]. CEA can also be easily extracted from sampling pads, and
quantication of the CEA concentration could add another layer of
informationthe retrospective conrmation of ROM with clear amniotic uid versus ROM with MSAF.
In conclusion, CEA could serve as a biomarker for MSAF. Elevated
levels of AFP and CEA in uid of unknown origin could help in the
conrmation of ROM with MSAF. A reliable bedside test for ROM and
ROM with MSAF, on the basis of sampling uids absorbed in pads,
could be a useful tool for midwives and obstetricians practicing in or
out of a hospital when encountering women with questionable ROM,
or when the typical green color of meconium is not clearly visible.
Conict of interest
The authors have no conicts of interest.
References
[1] Mor A, Tal R, Haberman S, McCalla S, Irani M, Perlman J, et al. Alpha-fetoprotein as a
tool to distinguish amniotic uid from urine, vaginal discharge, and semen. Obstet
Gynecol 2015;125(2):44852.
[2] Sarandakou A, Protonotariou E, Rizos D. Tumor markers in biological uids associated
with pregnancy. Crit Rev Clin Lab Sci 2007;44(2):15178.
[3] Sarandakou A, Kontoravdis A, Kontogeorgi Z, Rizos D, Phocas I. Expression of CEA,
CA-125 and SCC antigen by biological uids associated with pregnancy. Eur J Obstet
Gynecol Reprod Biol 1992;44(3):21520.
[4] Lolis D, Baltoyiannis P, Messinis IE. Carcinoembryonic antigen in high risk pregnancies. Int J Gynecol Obstet 1984;22(1):59.
[5] Roehr CC, Hansmann G, Hoehn T, Bhrer C. The 2010 Guidelines on Neonatal
Resuscitation (AHA, ERC, ILCOR): similarities and differenceswhat progress has
been made since 2005? Klin Padiatr 2011;223(5):299307.
[6] ACOG Committee Obstetric Practice. ACOG Committee Opinion Number 346, October
2006: amnioninfusion does not prevent meconium aspiration syndrome. Obstet
Gynecol 2006;108(4):1053.
[7] Xu H, Wei S, Fraser WD. Obstetric approaches to the prevention of meconium aspiration syndrome. J Perinatol 2008;28(Suppl. 3):S148.
[8] Sheiner E, Hadar A, Shoham-Vardi I, Hallak M, Katz M, Mazor M. The effect of meconium on perinatal outcome: a prospective analysis. J Matern Fetal Neonatal Med
2002;11(1):549.
[9] Barnhart KT. Novel diagnostic tests of ectopic pregnancy, if at rst you dont succeed.
Am J Obstet Gynecol 2015;212(1):46.