European Journal of Obstetrics & Gynecology and Reproductive Biology 176 (2014) 64–67

Contents lists available at ScienceDirect

European Journal of Obstetrics & Gynecology and

Reproductive Biology
journal homepage: www.elsevier.com/locate/ejogrb

The role of C-reactive protein measurement as a diagnostic aid in early

pregnancy
Yoni Cohen, Jessica Ascher-Landsberg, Aviad Cohen, Joseph B. Lessing, Dan Grisaru *
Department of Obstetrics and Gynecology, Lis Maternity Hospital, Tel Aviv Sourasky Medical Center, Affiliated to the Sackler Faculty of Medicine, Tel Aviv
University, 6 Weizmann Street, Tel Aviv 64239, Israel

A R T I C L E I N F O A B S T R A C T

Article history: Objective: To investigate the values of C-reactive protein (CRP) in early pregnancy for detection of any
Received 1 December 2013 abnormality of the conceptus and especially the condition of extra-uterine pregnancy (EUP).
Received in revised form 19 February 2014 Study design: In this prospective observational study, we established reference values for CRP in early
Accepted 3 March 2014
pregnancy. Next, we tested whether a single, wide-range CRP measurement could serve as a diagnostic
tool for abnormal first trimester pregnancy by comparing CRP levels in three different pregnancy
Keywords: statuses: viable intrauterine pregnancy (Group A), EUP (Group B) and delayed abortion (Group C).
Abortion
Results: CRP levels were significantly higher in normal pregnancy (Group A) compared to abnormal
C-reactive protein
pregnancy (Group B + C and Group B alone). CRP level was influenced only by the pregnancy status
Ectopic pregnancy
Pregnancy (normal, EUP or delayed abortion) and not by age, BMI, hematocrit or gravidity. The multiple logistic
regression model (adjusted for age, gravidity, gestational age, hematocrit and BMI) revealed CRP as being
a predictor for normal intrauterine pregnancy.
Conclusion: This study examined the association between CRP levels and abnormal first trimester
pregnancies. Our results support single CRP measurement as a diagnostic tool in early pregnancy.
ß 2014 Elsevier Ireland Ltd. All rights reserved.

1. Introduction
A pregnant state can be confirmed shortly after a missed
menstrual period, based on the sensitive pregnancy tests currently
available. This knowledge may well precede the detection of the
clinical or ultrasonographic signs of conception. The question of
the normality of the conceptus is inevitably raised, and the search
for diagnostic tools has led to the testing of a number of
biochemical markers, including serum creatine kinase, pregnan-
cy-associated protein A (PAPP-A) and relaxin [1–3]. The low levels
of accuracy of these markers, however, precluded their clinical use.
Early information about pregnancy status is even more relevant for
extra-uterine pregnancies (EUPs), where early diagnosis can
prevent clinical catastrophes and may allow conservative rather
than surgical management.
The immune system plays a central role in the normal
physiologic response to pregnancy, and the maternal inflammato-
ry response is already present from the early stages, but there are
few published data on the inflammatory response to early
abnormal pregnancy. Pro-inflammatory mediators, macrophages
* Corresponding author. Tel.: +972 3 692 5622; fax: +972 3 692 5670.
E-mail address: grisaro@post.tau.ac.il (D. Grisaru).
and natural killer cells are activated and play a part in the normal
implantation of human pregnancy [4,5]. C-reactive protein (CRP) is
an acute-phase reactant and one of the most important markers of
inflammation. CRP is produced by hepatocytes in response to
increased levels of interleukin-1 (IL-1) and IL-6. It is widely used in
medicine for the detection of inflammatory and infectious
conditions. Theron et al. [6] reported that there was a difference
in CRP levels between EUP and acute infection. We hypothesized
that the physiologic inflammatory response to abnormal pregnan-
cy might be altered, and therefore we designed the present study to
investigate the values of CRP in early pregnancy for the purpose of
detecting abnormality of the conception and to set reference
values that will aid in the differential diagnosis of early pregnancy
infection and inflammatory conditions.
2. Materials and methods
2.1. Study population
This prospective observational study was conducted at the Lis
Maternity Hospital of the Tel Aviv Medical Center following
Institutional Review Board approval. Eighty-nine women were
enrolled in this study. All the participants signed an informed
consent and were interviewed by a senior physician. The inclusion

http://dx.doi.org/10.1016/j.ejogrb.2014.03.004
0301-2115/ß 2014 Elsevier Ireland Ltd. All rights reserved.
 Y. Cohen et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 176 (2014) 64–67
criteria were: patients with known first trimester pregnancy (up
to 14 weeks of pregnancy) who were referred to our department
with abdominal pain, vaginal bleeding and suspected EUP; normal
pregnancy participants were recruited sequentially. The exclu-
sion criteria were: pregnant women who had already been treated
with methotrexate for ectopic pregnancy and women with a
known chronic or acute inflammatory condition (e.g. inflamma-
tory bowel disease, or arthritis) or steroid/NSAID users. The
laboratory results were not used in determining the clinical
management of the participants and the providers were blinded to
the CRP levels.
2.2. Patient data and laboratory variables
Demographic and clinical data included age, gravidity, current
body mass index (BMI), current diseases, medications and recent
febrile illness. A single blood sample for CRP measurement was
obtained at admission (prior to final diagnosis) and processed
immediately. Wide-range CRP was analyzed by immunoturbidi-
metric assay on an ADVIA 1650, using a Bayer ADVIA kit for CRP
(Bayer, Leverkusen, Germany), which has a wide range of
sensitivity and a low limit of detection (0.12 mg/L). The wide-
range CRP reagent is a suspension of uniform polystyrene latex
particles coated with anti-CRP antibody. Agglutination takes place
when serum is added to the solution. The increased turbidity is
measured at 571 nm. CRP concentration is determined by a
calibration curve [7].
2.3. Clinical outcome and group assignment
Beta-human chorionic gonadotropin (beta-hCG) levels, ultra-
sonographic findings and surgery reports were used for the final
diagnosis. A diagnosis of EUP, in which the embryo implants in
the fallopian tube, was based on an inappropriate increase in
beta-hCG and one the following: a sonographic finding of
extrauterine gestational sac or a surgery report confirming tubal
pregnancy. Delayed abortion or miscarriage is defined as non-
viable intra-uterine pregnancy. It was diagnosed when the
transvaginal ultrasound scan showed an intrauterine gestational
sac containing a fetal pole with a crown-rump length (CRL)
of more than 6 mm with no cardiac activity. The presence of an
intrauterine gestational sac containing a fetal pole with cardiac
activity confirmed the diagnosis of a normal viable intrauterine
pregnancy (IUP). The reference range for CRP at early pregnancy
was determined in women with a documented viable IUP.
Following final diagnosis, the participants were divided into
three study groups: normal viable IUP (Group A), EUP (Group B),
delayed abortion (Group C).
2.4. Statistical analysis
Comparison between groups regarding continuous clinical
factors was performed using one-way ANOVA and the non-
parametric analysis of variance Kruskal–Wallis test, while the chi-
square test was used to assess the relations between group and
categorical variables. Pearson’s correlation coefficient was used to
examine the relations between beta-hCG and log-CRP.
Moreover, in order to examine which parameters are indepen-
dently associated with normal pregnancy, a multiple logistic
regression model was applied. A multiple linear regression analysis
was also performed including: age, gravidity, BMI, group,
hematocrit and gestational age, to evaluate the effect of each
one of these factors simultaneously on log-CRP.
Finally we used a receiver operating characteristic (ROC) curve
to assess the optimal cut-off value with the highest sensitivity and
specificity of log-CRP for defining a normal pregnancy.
65
Table 1
Demographic and clinical data.
Group A Group B Group C
Age (years) 29.03  7.61 31.97  5.35 32.19  4.78
Gravidity 2.55  1.57 2.67  1.72 2.38  1.56
Gestational age (weeks) 10.32  3.41* 6.64  1.38 8.31  1.49
Body mass index 22.56  4.68 22.79  3.88 21.95  3.98
Hematocrit (%) 36.82  2.51 36.91  2.77 36.59  2.15
Group A, normal intrauterine pregnancy; Group B, extra uterine pregnancy; and
Group C, delayed abortion. Data presented as mean  standard deviation.
*
P < 0.05.
Descriptive statistics are given as mean  standard error of the
mean (SEM). A P value <0.05 was considered significant. Box-and-
whisker plots are presented with the 25% and 75% percentile ranges
(box depth) and the T-bars indicating variability.
All statistical analysis was performed using SAS 9.2 for
Windows.
3. Results
A total of 89 women were enrolled in this study: 31 in Group A,
36 in Group B and 22 in Group C. Except for gestational age at
which CRP levels were drawn, the groups were similar in their
general demographic and clinical data (Table 1).
The CRP levels were 7.67  1.81 mg/L for Group A,
2.43  0.61 mg/L for Group B, and 5.08  1.66 mg/L for Group C
(Fig. 1). CRP levels were not normally distributed, and so comparison
of CRP values between the groups and regression analyses were done
after logarithmic transformation. Log CRP level was significantly
higher in Group A compared to Groups B + C (P = 0.0054) and to Group
B alone (P = 0.0006), but there was no statistically significant
difference when Group A was compared to Group C alone (P = 0.09).
The correlations between CRP and gravidity, age, beta-hCG,
hematocrit and BMI during pregnancy did not reach a level of
significance for each group analyzed separately or when all three
study groups were pooled together. Furthermore, we did not find
any correlation between CRP levels and the duration of pregnancy.
We compared hematocrit levels between our study groups, as a
patient’s plasma expansion or contraction might affect CRP
concentration, but no differences in hematocrit levels were found.
After adjustment for possible confounding factors using
multiple linear regression analysis, it emerged that the CRP level
was influenced only by the pregnancy status (normal, EUP or
delayed abortion) and not by age, BMI, gravidity or hematocrit. The
multiple logistic regression model (adjusted for age, gravidity,
gestational age, hematocrit and BMI) revealed CRP as being a
predictor for normal intrauterine pregnancy (Table 2). The
[(Fig._1)TD$IG]sensitivity and specificity of a single CRP measurement to diagnose
Fig. 1. Maternal serum C reactive protein at early pregnancy states. Box plot
diagram showing C reactive protein level measured in women with normal
pregnancy, extra-uterine pregnancy (EUP) and missed abortion.
66 Y. Cohen et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 176 (2014) 64–67

Table 2 (median CRP 3.68 mg/L and 1.49 mg/L, respectively) [5]. Theron
Odds ratio and 95% confidence limits for possible predictors of normal pregnancy.
et al. [6] compared CRP levels in patients with EUP to those in
Odds ratio 95% confidence limits P value patients with acute infectious processes. The CRP levels in women
Log CRP 1.62 a
1.05–2.50 0.029 with EUP (median 8 mg/L) were lower than the CRP levels in
Gestational age (weeks) 2.07 1.32–3.23 0.001 patients with an acute infectious process (175 mg/L), but those
Age (years) 0.89 0.78–1.00 ns authors did not compare the CRP levels in normal pregnancy to
Gravidity 1.34 0.82–2.17 ns those in abnormal pregnancy.
Body mass index 1.02 0.86–1.21 ns
Variations in CRP during the first trimester were not thoroughly
The probability of having a normal pregnancy (study Group A) vs. abnormal one investigated and changes in CRP levels throughout pregnancy are
(study Groups B + C). ns, non significant.
a
not consistent in different reports [11]. CRP is widely used in
For every increase in log CRP unit.
numerous clinical situations in obstetrics and gynecology. CRP
levels have been correlated with the severity of pre-eclampsia [12–
[(Fig._2)TD$IG] 14], chorioamnionitis [15], pre-term delivery [16–18] and for an
assessment of the severity of pelvic inflammatory disease.
Sensitivity According to our results, CRP levels in early normal pregnancies
1.0
0.9
are minimally increased. This represents the physiologic activation
0.8
of the immune system, but these values are well below those of
0.7 inflammatory processes. Using a highly sensitive, wide-range CRP
0.6 analyzer, we recorded lower levels of CRP in EUP and delayed
0.5 abortion than in normal pregnancy, which can be explained by the
0.4 altered physiologic immune response in these pathological
0.3 conditions.
0.2 The inflammatory response in early pregnancy is composed of
0.1 systemic and local adaptation of the immune system to pregnancy,
0.0 and is present early in the first trimester [5,19]. These changes
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 consist of an elevated peripheral white blood cell count and the
1 - Specificity activation of monocytes and granulocytes (innate immune
Fig. 2. Receiver operating characteristic curve for CRP. Receiver operating system). Suppression of T-helper and T-cytotoxic cells, combined
characteristic (ROC) curve in the diagnosis of normal pregnancy. with reduced secretion of IL-2, interferon-g and tumor necrosis
factor-b and the up-regulation of Th2 cells, results in increased
levels of IL-4, IL-6 and IL-13 [20–22]. The secretion of IL-6 with a
normal pregnancy were calculated using an ROC curve (Fig. 2). The synergistic effect of IL-1 promotes the production of CRP in
area under the curve derived from the ROC analysis was 0.745. For hepatocytes. CRP is an acute phase reactant playing a central role in
example, a CRP level of 1.92 mg/L has a sensitivity of 76% and a the immunological system. It activates the complement system,
specificity of 64% for diagnosing normal pregnancy; a CRP level of modulates neutrophils, and enhances the adhesion and recruit-
6 mg/L has a sensitivity of 44% with a specificity of 84%. As ment of monocytes [23].
expected, the higher the measured CRP level, the greater the Since all cases were referred to our tertiary medical center after
specificity for a diagnosis of normal pregnancy, on account of lower initial evaluation in the community, the prevalence of EUP and
sensitivity. delayed abortion in this study does not represent the true
prevalence in our general population. Our results were based on
4. Comment a single CRP measurement. We do not know what the changes of
the CRP levels were, compared to pre-conception levels, nor do we
The results of this prospective observational study revealed that have serial measurements during early pregnancy, which might
the plasma level of the acute phase protein, CRP, increases during increase the accuracy of this tool. Nevertheless, even a single CRP
the first trimester of pregnancy above the normal range for adults measurement supported the application of CRP in the early
(>2 mg/L) [8,9]. The magnitude of this acute phase response varies diagnosis of abnormal pregnancy. To adjust for confounding by the
between normal and abnormal pregnancies: CRP level is signifi- earlier gestational age at which CRP levels were drawn in the EUP
cantly higher in a normal pregnancy status compared to an group, we performed multiple linear regression analysis. This
abnormal one, such as EUP and delayed abortion. It is noteworthy analysis clearly showed that the CRP level is an independent
that this difference is more obvious when we compared CRP levels predictor of pregnancy status in the first trimester of pregnancy.
in normal pregnancy to those of EUP, but less evident when Our results show that a single CRP measurement, taken
compared to the levels in delayed abortion. This may be related to a between 5 and 14 weeks of gestation, could serve as an additional
wider range of CRP levels in the normal pregnancy and delayed tool for the diagnosis and management of EUP or otherwise
abortion groups, when compared to a solid range in the EUP group. complicated early pregnancy. In addition, mean plasma CRP levels
After adjusting for age, gestational age, gravidity and BMI, an measured in this study can be used as a diagnostic aid in the
elevated CRP level continued to be associated with a normal IUP. differential diagnosis of pregnancy state, infections and inflamma-
The results of this study are in accordance with previous studies tory conditions. Clinical management of early pregnancies will
reporting elevated inflammatory response in pregnancy, but to the continue to be based on serial beta-hCG measurements and
best of our knowledge this is the first study to focus on CRP levels ultrasound imaging. Nevertheless, accurate diagnosis of abnormal
as a diagnostic tool differentiating between normal and abnormal early pregnancy status when the sonographic findings are
early pregnancy. In an earlier investigation into CRP in pregnancy, equivocal is challenging, especially given the life-threatening
Watts et al. [10] reported that CRP levels starting from 22 weeks of potential of EUP. Moreover, ultrasonographic diagnosis of EUP
gestation were significantly higher compared to a non-pregnant requires expertise and equipment that are not always available:
population (7–9 mg/L for pregnant women not in labor). Studies in thus an easily measurable biological marker would be extremely
women undergoing IVF treatment showed higher CRP levels in valuable. Future research is needed to validate our findings and
pregnant women compared to those who were not pregnant clinical recommendations.
 Y. Cohen et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 176 (2014) 64–67
Our results of significantly increased CRP levels in normal
pregnancy and a clear association between CRP and normal
pregnancy, support the clinical application of this diagnostic tool in
early pregnancy, especially in combination with serial beta-hCG
measurements.
Acknowledgments
Robbie Dolev and Esther Eshkol are thanked for editorial
assistance.
References
[1] Lavie O, Beller U, Neuman M, Ben-Chetrit A, Gottcshalk-Sabag S, Diamant YZ.
Maternal serum creatine kinase: a possible predictor of tubal pregnancy. Am J
Obstet Gynecol 1993;169:1149–50.
[2] Witt BR, Wolf GC, Wainwright CJ, Johnston PD, Thorneycroft IH. Relaxin, CA-
125, progesterone, estradiol, Schwangerschaft protein, and human chorionic
gonadotropin as predictors of outcome in threatened and nonthreatened
pregnancies. Fertil Steril 1990;53:1029–36.
[3] Meunier K, Mignot TM, Maria B, Guichard A, Zorn JR, Cedard L. Predictive value
of the active renin assay for the diagnosis of ectopic pregnancy. Fertil Steril
1991;55:432–5.
[4] Smarason AK, Gunnarsson A, Alfredsson JH, Valdimarsson H. Monocytosis and
monocytic infiltration of decidua in early pregnancy. J Clin Lab Immunol
1986;21:1–5.
[5] Sacks GP, Seyani L, Lavery S, Trew G. Maternal C-reactive protein levels are
raised at 4 weeks gestation. Hum Reprod 2004;19:1025–30.
[6] Theron GB, Shepherd EG, Strachan AF. C-reactive protein levels in ectopic
pregnancy, pelvic infection and carcinoma of the cervix. S Afr Med J
1986;69:681–2.
[7] Arbel Y, Eros Y, Rogowski O, et al. Comparison of values of wide-range C-
reactive protein to high-sensitivity C-reactive protein in patients undergoing
coronary angiography. Am J Cardiol 2007;99:1504–6.
[8] McPherson RA, Pincus MR, Henry JB. Henry’s clinical diagnosis and manage-
ment by laboratory methods. 21st ed. Philadelphia: Saunders Elsevier; 2007.
67
[9] Gabay C, Kushner I. Acute-phase proteins and other systemic responses to
inflammation. New Engl J Med 1999;340:448–54.
[10] Watts DH, Krohn MA, Wener MH, Eschenbach DA. C-reactive protein in normal
pregnancy. Obstet Gynecol 1991;77:176–80.
[11] Belo L, Santos-Silva A, Rocha S, et al. Fluctuations in C-reactive protein
concentration and neutrophil activation during normal human pregnancy.
Eur J Obstet Gynecol Reprod Biol 2005;123:46–51.
[12] Hwang HS, Kwon JY, Kim MA, Park YW, Kim YH. Maternal serum highly
sensitive C-reactive protein in normal pregnancy and pre-eclampsia. Int J
Gynaecol Obstet 2007;98:105–9.
[13] Kumru S, Godekmerdan A, Kutlu S, Ozcan Z. Correlation of maternal serum
high-sensitive C-reactive protein levels with biochemical and clinical
parameters in preeclampsia. Eur J Obstet Gynecol Reprod Biol
2006;124:164–7.
[14] Thilaganathan B, Wormald B, Zanardini C, Sheldon J, Ralph E, Papageorghiou
AT. Early-pregnancy multiple serum markers and second-trimester uterine
artery Doppler in predicting preeclampsia. Obstet Gynecol 2010;115:
1233–8.
[15] Trochez-Martinez RD, Smith P, Lamont RF. Use of C-reactive protein as a
predictor of chorioamnionitis in preterm prelabour rupture of membranes: a
systematic review. BJOG 2007;114:796–801.
[16] Lohsoonthorn V, Qiu C, Williams MA. Maternal serum C-reactive protein
concentrations in early pregnancy and subsequent risk of preterm delivery.
Clin Biochem 2007;40:330–5.
[17] Scholl TO, Chen X, Goldberg GS, Khusial PR, Stein TP. Maternal diet, C-reactive
protein, and the outcome of pregnancy. J Am Coll Nutr 2011;30:233–40.
[18] Bullen BL, Jones NM, Holzman CB, et al. C-reactive protein and preterm
delivery: clues from placental findings and maternal weight. Reprod Sci
2013;20:715–22.
[19] Sacks G, Sargent I, Redman C. An innate view of human pregnancy. Immunol
Today 1999;20:114–8.
[20] Chard T, Craig PH, Menabawey M, Lee C. Alpha interferon in human pregnancy.
BJOG 1986;93:1145–9.
[21] Veenstra van Nieuwenhoven AL, Heineman MJ, Faas MM. The immunology of
successful pregnancy. Hum Reprod Update 2003;9:347–57.
[22] Germain SJ, Sacks GP, Sooranna SR, Sargent IL, Redman CW. Systemic inflam-
matory priming in normal pregnancy and preeclampsia: the role of circulating
syncytiotrophoblast microparticles. J Immunol 2007;178:5949–56.
[23] Ablij H, Meinders A. C-reactive protein: history and revival. Eur J Intern Med
2002;13:412.