Correspondence: Professor M. F. Aksu, Department of Obstetrics and Gynaecology, Cerrahpasa Medical Faculty, 34301 Istanbul, Turkey.
Objective To determine the correlation between placental bed biopsy findings, and blood levels of the vascular cell adhesion molecule (VCAM-1) and fibronectin in pre-eclampsia/eclampsia, and to evaluate the relationship between these variables and severity of the disease.
Design A prospective case-control study.
Setting Department of Obstetrics and Gynaecology, Cerrahpasa Medical Faculty, Istanbul.
Sample Thirty-five women with pre-eclampsia/eclampsia were compared with 25 healthy women with uncomplicated pregnancies. Twenty-one placental bed biopsies from pre-eclamptic pregnancies were compared with 17 from normal pregnancies.
Main outcome measures Peripheral venous blood levels of VCAM-1 and fibronectin, measured by enzyme-linked immunoassay and radial immune diffusion technique.
Results In pre-eclampsia/eclampsia, blood VCAM-1 and fibronectin levels were higher than in normal pregnancy. Levels of both fibronectin and VCAM-1 correlated significantly with the diastolic blood pressure (r= 0.49 and r= 0.65, P < 0.001). There was also a significant direct linear correlation between plasma fibronectin and VCAM-1 levels (r= 0.57, P < 0.001). Normotensive women all had normal placental bed biopsy findings and the incidence of pathological placental bed biopsy findings increased with the severity of the pre-eclampsia.
Conclusion Inadequate trophoblastic invasion of spiral arteries, and elevated levels VCAM-1 and fibronectin were found in women with pre-eclampsia. The magnitude of defective trophoblastic invasion, and blood levels of VCAM-1 and fibronectin correlate with the clinical severity of pre-eclampsia.
Pre-eclampsia, which complicates approximately 7% of all pregnancies, remains a major cause of maternal and perinatal morbidity and mortality1. Pre-eclampsia is clearly a complex clinical syndrome potentially involving all of the organ systems. Despite decades of intensive research, the pathogenesis of pre-eclampsia remains incompletely understood. Placental ischaemia, immune maladaptation and genetic factors are probably all involved to some extent2.
Establishment of the human placenta requires that fetal trophoblast stem cells in the anchoring chorionic villi become invasive. This trophoblast aggregates into cell columns and invades both the uterine interstitium and vasculature, anchoring the fetus to the mother and establishing blood flow to the placenta. In normal pregnancy, trophoblast colonising spiral arterioles replaces maternal endothelium as far as the inner third of the myometrium. Placental bed biopsy studies have shown that the basic lesion in the uteroplacental bed, seen in pre-eclampsia, is a lack of or an incomplete invasion of trophoblasts into the maternal spiral arteries3–5.
Increasing evidence suggests that generalised endothelial cell damage and dysfunction are major pathophysiologic features of pre-eclampsia6. Plasma fibronectin levels as an indicator of endothelial cell injury have been shown to increase in pre-eclampsia7–9. This increase is found even before the clinical onset of the syndrome and plasma fibronectin measurement may be a useful prediction test7,10.
Although, the mechanisms of endothelial activation and dysfunction are yet unclear, there is substantial evidence suggesting that leukocyte-instigated damage contributes to this process11,12. It is believed that the vascular cell adhesion molecule (VCAM-1) can be utilised as a marker to monitor endothelial and leukocyte activation and has been shown to increase in the plasma of pre-eclamptic patients13–15.
In the present study we obtained placental bed biopsies and measured circulating VCAM-1 and fibronectin levels in normal pregnant women and pre-eclampsia. The purpose of the study was to determine the correlation between VCAM-1 and fibronectin levels and also to assess the correlation of these two markers with the severity of pre-eclampsia. We also evaluated the relation between the placental bed biopsy findings with these two markers and the severity of the disease.
Informed consent was obtained from all women. Approval for the study was given by the ethics committee of our hospital. The study population consisted of 25 women with normal pregnancy and 35 women with pre-eclampsia/eclampsia who were diagnosed and treated in our department. The hypertensive women were subclassified as mild pre-eclampsia (n= 16), severe pre-eclampsia (n= 16) and eclampsia (n= 3). Pre-eclampsia was defined as the onset of hypertension (blood pressure of ≥ 140/90 mmHg, ≥ 6 hours apart) and consistent proteinuria (300 mg / day or more) during the latter half of the pregnancy and both remitting remotely after delivery. The criteria for distinguishing severe pre-eclampsia from mild pre-eclampsia were a blood pressure > 160/110 mmHg or proteinuria > 3 g / day. In addition, any woman with cerebral or visual disturbances, epigastric pain, pulmonary oedema, oliguria (400 mL or less in 24 hours), or an abnormal platelet count and liver function profile was included in the severe pre-eclampsia group. Eclampsia was defined as the development of convulsions in patients with signs and symptoms of pre-eclampsia in the absence of other causes of convulsions. No cases of chronic hypertension or superimposed pre-eclampsia were included in this study. Other exclusion criteria were the presence of active infectious disease or medical complications including autoimmune disorders, diabetes mellitus and inflammatory conditions. Diastolic blood pressure was taken with a cuff sphygmo-manometer using the fifth Korotkoff sound. The average of repeated diastolic blood pressure measurements was taken into consideration. The presence of proteinuria was screened using dipsticks, and the amount of protein was measured in a 24-hour urine sample.
Twenty-five healthy pregnant women served as controls. None had signs of elevated blood pressure or other pregnancy complications and all gave birth to healthy infants. The indications for caesarean section in this group were tried to be at equivalent gestational ages with the hypertensives, and were for presentation abnormalities or previous caesarean section. These indications were selected to exclude some other pathologies for which we could not control.
Single peripheral venous blood samples were obtained from all women included in the study. Blood samples from pre-eclamptic and eclamptic women were drawn during labour or within two weeks of delivery between 28 and 40 weeks as soon as the diagnosis was established. Samples from the normotensive group were obtained between 26 and 40 weeks of gestation. The samples were stored at −30C° until the analyses were performed. Plasma levels of human soluble VCAM-1 (h-sVCAM-1) were quantified by an ELISA technique according to the manufacturer's instructions (Boehringer Mannheim, Germany). Plasma levels of fibronectin were detected by radial immunediffusion technique using LC partigen fibronectin (Boehringwerke AG, Marburg, Germany).
Placental bed biopsies were obtained at caesarean section from normal pregnancies (n= 17) and pre-eclamptic pregnancies (n= 21). Of the pre-eclamptic group, five were from mild pre-eclampsia, 13 were from severe pre-eclampsia and three were from eclampsia cases. At the time of caesarean section with direct visualisation of the placental site, two or three placental bed biopsies of at least 1 cm3 were taken with a scalpel and/or a pair of scissors. The specimens were fixed in buffered formalin. After fixation, dehydration and embedding in paraffin wax, 3 μm serial sections were cut and stained with haematoxylin and eosin. Specimens were examined by one of the authors (Z.C.) by light microscopy. Biopsies were included in the analysis if interstitial trophoblast and at least one spiral artery were present. Placental bed biopsies were classified as physiologic when the morphologic features of the normal human placental bed were documented16 or pathologic when there were inadequate endovascular trophoblast invasion and other well defined findings found in pre-eclampsia3.
The data were tested for normal distribution and Student's t test, Mann-Whitney U test or Fisher's exact test was used for comparison of two groups where appropriate. For comparison of more than two groups, Kruskal-Wallis one way analysis was performed and than nonparametric Scheffe test was utilised for evaluating group differences. Pearson's correlation coefficient was determined by linear regression analysis.
The clinical features of hypertensive and normotensive women are shown in Table 1. Mean week of delivery and birthweight were significantly lower (P<0.05, P<0.001), and intrauterine growth retardation and perinatal mortality rates were significantly higher (P<0.001) in hypertensive women. All of the eight perinatal mortality cases were growth retarded and <1500 g birthweight.
Table 1. The clinical features of hypertensive and normotensive women. Values are given as mean (SD) or n/n [%]. IUGR = intrauterine growth retardation.
Hypertensive (n= 35)
Normotensive (n= 25)
Diastolic blood pressure (mmHg)
Gestational age at birth (weeks)
The mean plasma levels of fibronectin and VCAM-1 and the incidence of pathologic bed biopsy findings in normotensive, mild and severe pre-eclamptic and eclamptic women are shown in Table 2. The mean plasma fibronectin and VCAM-1 levels of the severe group (pre-eclampsia and eclampsia cases considered together) were significantly higher than normotensive and mild pre-eclamptic women (P < 0.001). Although fibronectin and VCAM-1 levels were elevated in mild pre-eclampsia compared with normal pregnant women, the difference was not statistically significant (P>0.05). Considering placental bed biopsy findings, in normal pregnancies there were no pathologic bed biopsy finding (trophoblast invasion was seen in 100% of women). However, in the hypertensive group the incidence of pathologic bed biopsy findings increased with the severity of the disease state.
Table 2. The mean plasma levels of fibronectin and VCAM-I and the incidence of pathologic bed biopsy findings in normotensive, mild and severe pre-eclamptic and eclamptic women. Values are given as mean (SD) or n/n [%].
Normotensive (n= 25)
Mild pre-eclampsia (n= 16)
Severe pre-eclampsia (n= 16)
Eclampsia (n= 3)
Severe group (n= 19)*
*Severe pre-eclampsia and eclampsia cases against together.
†P < 0.oo1, values are significanlty higher in severe group than mild pre-eclampsia and normotensive cases.
Pathologic bed biopsy
The relation between fibronectin and VCAM-1 with diastolic blood pressure including both normotensive and pre-eclamptic women is shown in Fig. 1. The plasma levels of both fibronectin and VCAM-1 were significantly increased (r= 0.49 and r= 0.65, P < 0.001) with the increments in diastolic blood pressure. There was also a significant direct linear correlation with the increments of plasma fibronectin and VCAM-1 levels (r= 0.57, P<0.001).
When we correlate pathologic bed biopsies with the two serum markers; 14 women with pathologic and 24 women with normal bed biopsies, the mean plasma fibronectin and VCAM-1 levels were significantly higher in women with pathologic bed biopsies (P<0.001) (Table 3). However, if we take into consideration only the hypertensive women, fibronectin and VCAM-1 levels did not significantly change between normal and pathologic bed biopsy findings (Table 3).
Table 3. Fibronectin and VCAM-1 levels of women with normal and pathologic bed biopsy findings (all women and only hypertensive women are ilustrated separately). Values are given as mean (SD).
Pathologic bed biopsy
Normal bed biopsy
Pre-eclampsia remains one of the major causes of maternal and perinatal mortality. Prevention requires knowledge of pathophysiolgical mechanisms of disease, early detection of an at-risk group of women and means for the prevention or palliation of the pathological process. Although the precise aetiology remains elusive, endovascular cytotrophoblast invasion in the spiral arteries and endothelial cell dysfunction are two key features in the pathophysiology of pre-eclampsia17.
Defective trophoblastic invasion is the single consistent histologic feature in pre-eclampsia5,16. The available epidemiological evidence supports the view that the origin of this placentation defect is multifactorial; immune maladaptation, genetic predisposition and intrinsic defect in differentiation of trophoblasts may all play a role2,18. Placental bed biopsy studies have shown that trophoblastic invasion is not an all or none phenomenon and there is a gradient in the percentage of arteries invaded from normal pregnancy to pre-eclampsia3,19. There are approximately 100 to 150 spiral arteries in the placental bed and it is impossible to examine them all. The one that we sampled was assumed to represent the whole. In our study, normotensive women all had physiological changes in the spiral arteries. Whereas in the pre-eclamptic women, the percentage of pathologic bed biopsies increased with the severity of the disease. This finding confirms the causal relationship between the magnitude of defective trophoblastic invasion and the clinical severity of the disease process.
Endothelial cells line all blood vessels and regulate thrombosis, vascular tone and the permeability of the vessel wall to cells and molecules via the synthesis and release of a number of diverse factors. There is now considerable evidence that damage and dysfunction of endothelial cells of maternal blood vessels is a key feature in the pathophysiological changes of preeclampsia17. Fibronectin as a potential marker for endothelial vascular injury has been found to increase in pre-eclampsia7–10. The findings of the present study also confirm those studies. The mechanisms of endothelial activation and dysfunction are yet unclear. However, there is substantial evidence suggesting that leukocyte instigated damage may contribute to this process, and VCAM-1 as a marker to monitor endothelial and leukocyte activation was found to be elevated in pre-eclampsia12–14. Consistent with previous reports, we observed significantly elevated levels of VCAM-1 in pre-eclamptic women. Serum concentrations of fibronectin and VCAM-1 correlate with disease severity and are significantly higher in severe pre-eclampsia/eclampsia compared with mild pre-eclampsia or healthy pregnant women. If we assume that the extent of endothelial activation and damage is more pronounced in severe disease and the prognosis of mild pre-eclampsia is relatively benign compared with severe pre-eclampsia and eclampsia, this is not a surprising finding.
We found a good correlation between diastolic blood pressure and plasma fibronectin and VCAM-1 levels. There was also a good correlation between fibronectin and VCAM-1 levels, fibronectin is found to significantly increase with increments in VCAM-1. To the best of our knowledge, this is the first study to show the relation between these two markers. The release of VCAM-1 correlates to plasma fibronectin concentrations, probably indicating a causal relationship. Taking into consideration VCAM-1, fibronectin and diastolic blood pressure findings together, we can speculate that leukocyte-instigated endothelial damage correlates with the clinical spectrum of pre-eclampsia syndrome.
Serum concentrations of VCAM-1 and fibronectin were significantly elevated in women with pathological bed biopsy findings. However, as all the pathological bed biopsies were from hypertensive cases, we separately looked at only the hypertensive women. Although serum VCAM-1 and fibronectin levels of hypertensive women with pathologic bed biopsies were elevated compared with the ones with physiological bed biopsies, the difference was not significant. This may be due to small number size or sampled spiral arteries may not be the representative of all the pathology in those pre-eclamptic women. We may also speculate that, inadequate trophoblastic invasion of spiral arteries is a unique phenomenon in pre-eclampsia but yet not the sole cause of leukocyte activation and endothelial damage seen in this disease.
The present study confirms the findings of inadequate trophoblastic invasion of spiral arteries, elevated VCAM-1 and fibronectin levels in pre-eclampsia. The magnitude of defective trophoblastic invasion, serum concentrations of VCAM-1 and fibronectin correlate with the clinical severity of the disease process. The release of VCAM-1 also correlates with plasma fibronectin concentrations. To our knowledge this is the first study to correlate placental bed findings with markers of both endothelial damage and the mechanism of neutrophil activation. The present study confirms a relationship between the pathological events in the placental bed, neutrophil activation and endothelial damage in pre-eclampsia.