Levels of C-reactive protein in pregnant women who subsequently develop pre-eclampsia


*Professor K. Nicolaides, Harris Birthright Research Centre for Fetal Medicine, King's College Hospital, Denmark Hill, London SE5 8RS, UK


Objective To investigate whether a maternal inflammatory response precedes the development of pre-eclampsia.

Design Cross-sectional study.

Setting Antenatal clinic in an inner city teaching hospital.

Population Two groups of women were examined at 23–25 weeks of gestation. The first group (45 women) had normal uterine artery Doppler waveforms and subsequently had a normal pregnancy outcome. The second group (45 women) had Doppler evidence of impaired placental perfusion and 21 (47%) of them had normal outcome, 14 (31%) developed intrauterine growth restriction and 10 (22%) developed pre-eclampsia, with or without intrauterine growth restriction.

Methods C-reactive protein, an acute-phase reactant, was measured in maternal serum using a highly sensitive method with a detection limit of 0.05mg/L.

Main Outcome Measures Development of pre-eclampsia, as defined by the International Society for the Study of Hypertension in Pregnancy. Intrauterine growth restriction was defined as birthweight <5th centile for gestation and sex of the neonate.

Results The serum C-reactive protein concentration in women who subsequently developed pre-eclampsia (median 1.56, range 0.55–3.12mg/L) or delivered a baby with birthweight <5th centile (median 0.74, range 0.64–1.58mg/L) was not significantly different from that in women with uncomplicated pregnancies (median 1.28, range 0.75–2.08mg/L; P= 0.95 and P= 0.62, respectively).

Conclusion These findings suggest that the onset of clinical signs of pre-eclampsia may not be preceded by a maternal inflammatory response, as assessed by measurement of C-reactive protein.


Pre-eclampsia is one of the most serious complications of pregnancy and one of the leading causes of maternal and perinatal morbidity and mortality1. The maternal syndrome is characterised by hypertension, proteinuria and, in its most severe form, is associated with thrombocytopenia, disseminated intravascular coagulation and hepatocellular damage. Many of these features are also found in the systemic inflammatory response syndrome, suggesting that pre-eclampsia may involve an excessive maternal inflammatory response to pregnancy. In accordance with this, several studies have shown that pre-eclampsia is associated with activation of complement, blood leucocytes and increased levels of various cytokines such as interleukin-6, which are suggestive of a generalised intravascular inflammatory reaction2–6. Many features of pre-eclampsia can be explained by a disruption of the normal vasodilator and antithrombotic function of the maternal endothelium7. Moreover, inflammation itself has recently been reported to induce endothelial dysfunction8.

The factors that initiate pre-eclampsia are unknown. However, it has been postulated that failure of trophoblastic invasion of the spiral arteries is the primary insult7,9. Doppler ultrasound has been used to assess the uteroplacental circulation and several studies have reported that high impedance in the uterine arteries during the second trimester of pregnancy, prior to the development of any clinical features of pre-eclampsia, is associated with adverse outcome (subsequent development of pre-eclampsia and/or intrauterine growth restriction)10,11. Furthermore, the presence of bilateral notching of the uterine arteries, a qualitative assessment of the flow velocity waveform, has been shown to be a highly sensitive predictor of pre-eclampsia and/or intrauterine growth restriction12.

The aim of the present study was to compare levels of C-reactive protein (as a marker of the systemic inflammatory response) in women at high and low risk of pre-eclampsia based on colour Doppler examination of the uterine arteries during the second trimester of pregnancy.


All women attending for routine antenatal care at King's College Hospital have colour Doppler examination of their uterine arteries at 23–25 weeks of gestation (Acuson Aspen, Acuson Co, Mountain View, California, US), performed by an experienced operator. The right and left uterine arteries are identified at the apparent cross-over with the external iliac arteries and pulsed wave Doppler is used to obtain waveforms. When three similar consecutive waveforms are obtained, the presence of an early diastolic notch is noted, the pulsatility index is measured and the mean pulsatility index of the two vessels is calculated.

This was a cross-sectional study involving 45 pregnant women with normal uterine artery Doppler waveforms and 45 pregnant women with abnormal uterine artery Doppler waveforms (presence of early diastolic notch bilaterally) at 23–25 weeks of gestation recruited from the antenatal clinic of King's College Hospital. All women had normal singleton pregnancies with fetuses of appropriate size for dates. All mothers were healthy with no medical problems, with no apparent signs of infection and were on no medication. At study entry gestational age at screening, maternal age, body mass index, ethnic group, smoking status, parity and heart rate were recorded. Blood pressure was measured in the right arm with the subject seated using an ambulatory blood pressure monitor (SpaceLabs Medical, 90,207 WA, USA). Three measurements were taken and averaged. The study was approved by the Local Ethics Committee and all subjects gave written informed consent.

Blood Sampling and C-Reactive Protein Measurement

Venous blood was drawn from each patient and centrifuged. Serum was then aliquoted and stored at −70°C. Samples were thawed at room temperature, vortexed and centrifuged prior to analysis. High sensitive C-reactive protein was assayed using an Immulite automated immunoassay analyser (DPC, Gwynedd, Wales), which employs chemiluminescence for detection. All samples were pre-diluted using diluent provided by the manufacturer according to their instruction. Samples (10μL) were pippetted from the sample cup into a test module cotaining a ligand-labelled anti-C-reactive protein MAb. An alkaline phosphatease labelled anti-C-reactive protein polyclonal antibody was then added. The test module contains an immobilised anti-ligand bead. This was incubated at 37°C for 30 minutes with automated intermittent agitation of the test module. During the incubation, the C-reactive protein and antibodies form an antigen–antibody sandwich complex which binds to the solid phase. A centrifugal washing step removes unbound conjugate. A phosphate ester of adamantyl dioxetane is added which undergoes hydrolysis in the presence of alkaline phosphatase, producing an unstable intermediate, which is continuously produces light emission. This is detected by the luminometer and is proportional to the C-reactive protein concentration of the sample. The high sensitive C-reactive protein assay coefficients of variations were 9.2% at 0.12mg/L, 5.3% at 0.99mg/L, 4.9% at 0.168mg/L and 5.6% at 0.330mg/L. The assay had a calibration range up to 50mg/L and a detection limit of 0.05mg/L. The samples were handled in identical and blinded fashion.

Defining Clinical Outcome

Pre-eclampsia was defined according to the criteria of the International Society for the Study of Hypertension in Pregnancy. Under this classification, hypertension is defined as one diastolic blood pressure reading ≥110 mmHg or two consecutive diastolic blood pressure readings ≥90 mmHg at least four hours apart. Significant proteinuria is defined ≥300 mg total protein in a 24-hour urine collection or, if this is not available, 1+ proteinuria by dipstick on two consecutive occasions at least four hours apart. Using this classification pre-eclampsia was defined as hypertension in combination with proteinuria developing after 20 weeks of gestation in a group of previously normotensive women. Intrauterine growth restriction was defined by a birthweight below the 5th centile for gestation and sex of the neonate13.

Statistical Analysis

Normality of the distribution of the data was examined with the Shapiro–Wilk test. For those parameters that were not normally distributed logarithmic transformation was performed. Descriptive data are expressed as mean (SD) or as median (interquartile range), where appropriate. The different parameters in pregnant women with normal uterine artery Doppler waveforms and normal outcome, bilateral notches and normal outcome, bilateral notches with subsequent development of intrauterine growth restriction and bilateral uterine artery notches with subsequent development of pre-eclampsia were compared using one-way analysis of variance followed by a post hoc test if appropriate. The χ2 test was used to compare categorical variables among groups. Univariate and multiple logistic regressions, using as dependent variable the levels of C-reactive protein, were performed to look for possible determinants of C-reactive protein.


None of the women with normal uterine artery Doppler waveforms developed pre-eclampsia and all of them delivered appropriately grown infants. The majority of these women had spontaneous onset of labour except two who had induction of labour due to postmaturity. Women with bilateral notches of the uterine arteries at 23–25 weeks (n= 45) were classified into three groups according to the outcome of pregnancy: those with no complications (n= 21, 47%); those who developed intrauterine growth restriction without pre-eclampsia (n= 14, 31%); and those who developed pre-eclampsia (n= 10, 22%), including four with intrauterine growth restriction. In the group of women with bilateral notches and normal outcome 18 had spontaneous onset of labour, two women had induction of labour at 41 weeks due to prolonged rupture of membranes and one had an elective caesarean section at 38 weeks on maternal request.

The demographic, clinical and biochemical characteristics of the women according to the outcome of pregnancy are given in Table 1 There was no statistically significant difference between the groups in baseline demographic characteristics. The mean (SD) of women who eventually developed pre-eclampsia had significantly higher systolic and diastolic blood pressure compared with those who did not (123.4mmHg (7.7) vs 115mmHg (8.5), P < 0.05 and 75.3mmHg (9) vs 66.5mmHg (6.6), P < 0.05, respectively). At delivery, the systolic and diastolic blood pressure of these women was 164mmHg (12) and 105mmHg (5), respectively. As expected, women who developed pre-eclampsia had significantly higher pulsatility index of the uterine arteries and delivered smaller fetuses earlier compared with the women who did not have any complications of pregnancy. The levels of high sensitive C-reactive protein were similar in pregnant women who subsequently developed pre-eclampsia and in women without complications of pregnancy (1.56mg/L (0.55–3.12) vs 1.28mg/L (0.75–2.08), P= 0.95) (Fig. 1). Of the group of women who developed pre-eclampsia, half required delivery before 34 weeks of gestation. The levels of C-reactive protein between those who delivered before and those who delivered after 34 weeks of gestation were not significantly different (1.61mg/L (0.64–5.23) vs 1.51mg/L (0.49–3.3), P= 0.69). Similarly, in the pre-eclampsia group there were no significant differences between those with and those without intrauterine growth restriction (1.68mg/L (0.5–6.5) vs 1.56mg/L (0.55–2.64), P= 0.83). Univariate and multiple regression analyses, using as dependent variable the levels of C-reactive protein and as possible determinants the maternal age, ethnic group, smoking status, parity, heart rate, blood pressure, presence or absence of bilateral notches and presence or absence of pre-eclampsia did not reveal any statistical significance.

Table 1.  Demographic, clinical maternal and neonatal clinical characteristics in each group of women. Values are shown as n (%), mean [SD] or median (range). All the comparisons were performed with the group of women without bilateral notches at examination. PI = pulsatility index; BMI = body mass index; PE = pre-eclampsia.
CharacteristicNo notches–Normal outcome (n= 45)Notches present–Normal outcome (n= 21)Notches present–Development of IUGR (n= 14)Notches present–Development of PE (n= 10)
  1. *P<0.05,

  2. P<0.0001.

Gestational age at screening (wk)24 (23.5–25)24 (24–25)24.5 (24–25)24 (24–25)
Maternal age (years)29 [5.5]26.3 [6.3]27 [6.9]26.9 [6]
Ethnic group (Caucasian)24 (53.3)9 (42.9)8 (57.1)6 (60)
BMI (kg/m2)26.44 (24.2–29.6)26.14 (22.9–30.2)25.45 (20.9–26.7)26.97 (25.1–30.5)
Smokers11 (24.4)6 (28.6)5 (35.7)2 (20)
Nulliparity25 (55.6)11 (52.4)13 (92.9)7 (70)
Heart rate (bpm)84.7 [10.1]77.3 [12.4]79.2 [10.5]78.9 [10]
Systolic BP (mmHg)114.7 [8.5]112.7 [9.2]114.2 [10.2]123.4 [7.7]*
Diastolic BP (mmHg)66.5 [6.6]64.1 [15.4]66.7 [5.75]75.3 [9]*
Mean PI of uterine arteries0.81 [0.24]1.41 [0.56]1.87 [0.64]1.65 [0.48]
Gestational age at delivery (wk)39.4 [1.6]40.2 [1.7]37.9 [4]34 [3.5]
Birthweight (g)3322 [501]3254 [400.8]2180.7 [733.6]2019.6 [706.6]
C-reactive protein (mg/L)1.28 (0.75–2.08)1.49 (0.68–1.87)0.74 (0.64–1.58)1.56 (0.55–3.12)
Figure 1.

Levels of C-reactive protein (C-reactive protein) in the different groups of pregnant women at 23–25 weeks of gestation (Group 1: no notches, normal outcome, Group 2: notches present, normal outcome, Group 3: notches present, subsequent delivery of a small for gestational age neonate, Group 4: notches present, subsequent development of pre-eclampsia). The vertical lines illustrate the median C-reactive protein levels in each group.


The study has shown that maternal inflammatory response, as assessed by measurement of C-reactive protein, is not different in women prone to pre-eclampsia before the onset of clinical symptoms compared with pregnant women with uncomplicated pregnancies.

Although the pathogenesis of pre-eclampsia is unknown, a generalised inflammatory process has recently been implicated in its pathogenesis5,14. Activated circulating leucocytes, increased production of reactive oxygen species, increased release of inflammatory cytokines, such as tumour necrosis factor-a and interleukin-6, abnormal activation of the clotting system in pre-eclamptic compared with normal pregnancies support this hypothesis2–6,15. Furthermore, Faas et al.16 have shown that a single microdose intravenous administration of endotoxin to rats at 14 days' gestation triggers hypertension and proteinuria that persists until the end of pregnancy. Additionally, postmortem observations on women with pre-eclampsia indicate that in some circumstances the lethal pathologic features resemble those of the Shwartzmam reaction, a particular form of inflammatory response to endotoxin in experimental animals17. Based on this evidence, Redman et al.14 have recently suggested that maternal endothelial dysfunction, which until now has been althought to play the primary role in the development of pre-eclampsia, is just a single component of a more widespread perturbation of all cells in the vascular compartment that comprises a generalised, intravascular, inflammatory response to pregnancy. However, in the majority of the previous studies inflammatory indices were measured when pre-eclampsia was already established making it difficult to understand whether systemic inflammation precedes and predicts the development of pre-eclampsia. In our study, the lack of difference in the levels of C-reactive protein between women who eventually developed pre-eclampsia, even the severe form of the disease requiring delivery before 34 weeks of gestation, and those who did not suggests either that the maternal inflammatory response in the second trimester of pregnancy is not exaggerated before the onset of the clinical signs of pre-eclampsia or that C-reactive protein is not part of the inflammatory response that might exist. But the fact that C-reactive protein levels, as measured by routine assays, have been reported to be increased in pregnant women with established pre-eclampsia suggests that the latter is unlikely and indicates that maternal inflammatory response does not precede the development of pre-eclampsia18,19.

A feature of most forms of inflammation and tissue damage is the increase in the circulating concentration of various proteins known as acute-phase reactants20. These reactants are mainly produced by hepatocytes and the increased expression of the acute-phase protein genes is driven by cytokines, which are produced by activated macrophages and other cells21. The concentration of C-reactive protein is a precise objective and sensitive, although non-specific, index of overall inflammatory activity in the body22. Newer highly sensitive assays for C-reactive protein have been developed that are able to detect low grade but clinically significant inflammation23. Using these methods C-reactive protein has been reported to be an important predictor of cardiovascular events both in the short and long term23–25. Furthermore, recent studies have suggested that increased levels of C-reactive protein are associated with an abnormal systemic endothelial vascular reactivity as assessed by forearm venous occlusion plethysmography26,27. During inflammation the concentration of C-reactive protein can rise by up to 10,000 fold22,28. It could be argued that the sample of our study was too small to investigate the hypothesis that significant inflammatory response precedes the manifestation of pre-eclampsia. However, with 10 subjects in each group our study had a more that 80% power (α= 5%) to detect a twofold increase29.

In order to identify women at risk of pre-eclampsia, we used uterine artery Doppler assessment during the second trimester of pregnancy. Many studies have evaluated the usefulness of different Doppler parameters in identifying high risk pregnancies and among them the presence of bilateral notches in the uterine artery flow velocity waveforms appears to be the best predictor of poor pregnancy outcome10–12. It has been reported that women with bilateral notches have about 25% chance of developing pre-eclampsia10,12 and this was the case in our population where almost a quarter of the women developed pre-eclampsia.

In summary, we have shown that pregnant women who subsequently develop pre-eclampsia have similar levels of C-reactive protein, as assessed by a highly sensitive method, to those who have uncomplicated pregnancies. These findings suggest that the onset of clinical signs of pre-eclampsia may not be preceded by a maternal inflammatory response.


The study was supported by The Fetal Medicine Foundation (UK Registered Charity number: 1037116). The authors would like to thank Dr D. Gaze (Cardiac Research Scientist, Clinical Biochemistry Department, St George's Hospital, London) for performing the analysis of high sensitive C-reactive protein.