Preeclampsia, a pregnancy-specific multiorgan disorder, has been proposed to be a two-stage disease in which abnormal placentation is followed by generalized endothelial cell dysfunction.1,2 It has been suggested that the hypoxic placenta releases factors into the maternal circulation that give rise to the endothelial cell dysfunction and the clinical features of the syndrome.2,3 Elucidation of these released factors and their actions may provide a plausible link between abnormal placentation and the clinical manifestations of preeclampsia
Clinically, preeclampsia is a heterogeneous disease, and major differences have been observed between early- and late-onset preeclampsia.4,5 Early-onset disease is usually more aggressive to the mother and the fetus. Early, but not late preeclampsia may involve the fetus with intrauterine growth restriction.6 Poor placentation, as estimated by second-trimester Doppler ultrasound examination of uterine arteries, has a closer association with early- than with late-onset preeclampsia.7 The two-stage theory is therefore more consistent with early-onset preeclampsia than late-onset disease,8 and it therefore seems important to consider the early- and late-onset forms as two different entities when investigating the pathophysiology of preeclampsia.
Accumulating evidence suggests that the relationship between vascular endothelial growth factor-A (VEGF-A), placental growth factor (PlGF), and their common receptor soluble fms-like tyrosine kinase 1 (sFlt1) (also known as soluble VEGF receptor-1) is important for the control of vasculogenesis, angiogenesis, and placental development during pregnancy.9–11 Soluble Flt1 acts as a strong inhibitor of angiogenic activity by binding to and inactivating the proangiogenic factors VEGF-A and PlGF.12 Low serum levels of free VEGF-A and PlGF and elevated levels of sFlt1 appear to antedate the onset of clinical signs of preeclampsia by some weeks.13–17 It has also been shown that administration of exogenous sFlt1 to rats induces hypertension, proteinuria, and renal damage resembling the human glomerular endotheliosis, which is pathognomonic for preeclampsia.9
In most previous studies of the association between angiogenic factors and preeclampsia, patients with early- and late-onset preeclampsia have not been separated into different study groups.18,19 The following studies did separate early from late disease although they did so after subgroup analysis. Levine et al13 and Chaiworapongsa et al20 reported greater changes in sFlt1 in early than in late disease. Robinson et al21 found lower serum levels of PlGF in early than in late preeclampsia, but no difference in sFlt1.
Our hypothesis was that early-onset preeclampsia is more associated with placental ischemia than late-onset disease, leading to more pronounced alterations in sFlt1, VEGF-A, and PlGF levels.
The aims of our study were, therefore, to estimate whether both early and late preeclampsia was associated with changes in the plasma levels of sFlt1, VEGF-A, and PlGF and whether such alterations were more pronounced in early- than in late-onset disease.
MATERIALS AND METHODS
The study protocol was approved by the local Ethics Committee of the Medical Faculty of Uppsala University, and informed consent was obtained from each patient included in the study.
All study patients were recruited from the University Hospital in Uppsala during the time period 2001–2005. Women with preeclampsia with an early or late onset were eligible for study inclusion. Preeclampsia was defined as new-onset hypertension (140/90 mm Hg or greater) observed on at least two separate measurements 6 hours or more apart, combined with proteinuria (2 or greater on a dipstick or in a 24-hour urine sample showing 300 mg/24 hour or greater). The group with early-onset preeclampsia consisted of pregnant women with preeclampsia diagnosed before gestational week 32 who were delivered prematurely because of preeclampsia. The group with late-onset preeclampsia consisted of pregnant women with preeclampsia diagnosed in gestational week 35 or later.
Three different control groups were included. 1) The nonpregnant group comprised healthy, nonpregnant women who were recruited among women visiting a reproduction center for infertility based on male or tubal factors. They later became pregnant and delivered successfully. 2) The early control group comprised healthy pregnant women in gestational weeks 24–32. These women were recruited during a routine visit to an antenatal clinic on gestational weeks 24–32. Only those whose pregnancy continued normally and resulted in a full-term delivery of a healthy child with normal weight were included in the study. 3) The late control group comprised healthy pregnant women delivering in gestational weeks 36–42. Women in this group could be delivered either by cesarean or vaginally. In the latter case they had contacted the hospital at term before the active phase of labor, and they were enrolled if it was estimated that the delivery would occur within a few days (eg, ruptured membranes). A planned vaginal delivery could be converted to cesarean delivery according to the practice at the clinic.
Only women with single pregnancies were included. Women with a concurrent diagnosis of an upper urinary tract infection, chronic hypertension (hypertension before pregnancy and persistently elevated blood pressure before the 20th week of gestation), diabetes mellitus, or pre-existing renal disease were not included.
Antihypertensive treatment was commenced in both groups of women with preeclampsia if the systolic blood pressure rose above 170 mm Hg, if the diastolic rose above 110 mm Hg, or if both conditions existed. Small and large for gestational age was defined as birth weight 2 standard deviations below and above the mean birth weight for gestational age according to the nationwide-used charts, based on Albertsson-Wikland et al.21
Plasma samples were collected from each subject on entry into the study. Samples from nonpregnant controls were taken during an appointment at the reproduction center. In women in the early control group, samples were taken during a routine visit to a midwife. Plasma samples from women with preeclampsia were collected at the first visit to the hospital, before the start of active labor. In late controls the plasma samples were collected during a visit to the hospital when it was estimated that a delivery would follow within 2–3 days or during a preoperative visit before a planned cesarean delivery.
After collection, the plasma samples were immediately put into a refrigerator, where they were kept from 20 minutes to 2 (in a few cases up to 4) hours before being centrifuged for 10 minutes at 1,500g. The samples were then stored at –70°C until analyzed.
Samples were analyzed for sFlt1, PlGF, and VEGF-A by using commercially available enzyme-linked immunosorbent assay (ELISA) kits (DVR100B, DPG00, and DVE00, R&D Systems, Minneapolis, MN). Briefly, the microtiter plates had been coated with monoclonal antibodies specific for the antigen in question, and the first step was to add standards and samples to the wells. During the following incubation period, the antigen present in the standards and samples became bound to the immobilized antibody. After a thorough wash, an enzyme-linked polyclonal antibody specific for each protein was pipetted into the wells, and after a second incubation and wash step, a substrate solution was added and color developed in proportion to the amount of antigen bound. The color development was subsequently stopped and the absorbance at 450 nm was measured in a SpectraMax 250 (Molecular Devices, Sunnyvale, CA). The results were calculated according to the manufacturer's recommendations. The interassay coefficient of variation was approximately 7%. The detection limit of the VEGF-A ELISA test was 15 pg/mL.
A sample-size estimation was made, with a wanted power of 90% and α of 0.05 for detecting a difference in mean sFlt1 of 5,000 pg/mL between early- and late-onset preeclampsia (estimated from a previous observation of sFlt1 values in preeclampsia18). This resulted in a needed sample size of 17 women in each group.
The Kolmogorov-Smirnov test was used to test the data for a normal distribution. The Kruskal-Wallis test was thereafter used for overall comparison of medians in VEGF-A. In cases where the level of VEGF-A was below the test's detection limit (15 pg/mL) the level was recorded as 15 pg/mL in the protocol. Mann-Whitney U tests were used to estimate whether there were any differences in medians in sFlt1 and PlGF values between the study groups. For continuous variables in Table 1, we used analysis of variance for overall comparisons of means and Tukey test for multiple pair-wise comparisons. When only one comparison was made, Student t test was used. Chi-square and Fisher exact tests were used for comparisons between categorical variables.
All significance tests were two-tailed. P≤0.05 was considered to denote a statistically significant difference. A Bonferroni correction of the significance level was made for multiple Mann-Whitney U tests and pair-wise comparisons of percentages. For pair-wise comparisons of sFlt1 and PlGF, P≤.017 was considered to denote a statistically significant difference, whereas for pair-wise comparisons of percentages, the limit was .025. All statistical analyses were performed with SPSS 12.0 (SPSS Inc, Chicago, IL) for Windows.
The study groups did not differ significantly from each other in the baseline characteristic maternal age. Parity was similar between the pregnant groups. Smoking habits did not differ between early controls and early preeclampsia or between late controls and late preeclampsia (Table 1). Length of gestation at delivery was, on average, 13 days shorter for the late-onset preeclampsia group than for their controls, and body mass index (BMI) was higher in the group with early-onset preeclampsia than in early controls (27 versus 23). All women had normal blood pressure in the first trimester, but women who later developed early-onset preeclampsia had significantly higher mean diastolic blood pressure compared with their controls (77 mm Hg versus 69 mm Hg).
Nonpregnant women had a median plasma level of sFlt1 of 48 pg/mL (Fig. 1A). Late controls had a higher median plasma level of sFlt1 (7,827 pg/mL) than early controls (886 pg/mL) (P<.001) (Fig. 1). Women with early-onset preeclampsia had a higher median plasma level of sFlt1 (37,700 pg/mL) than early controls (P<.001). Women with late-onset preeclampsia had a higher median plasma level of sFlt1 (26,106 pg/mL) than late controls (P<.001). The relative increase, compared with the respective control group, was greater in women with early-onset (43 times) than in those with late-onset (3 times) preeclampsia.
Nonpregnant women had a median plasma level of PlGF of 110 pg/mL (Fig. 1B). The median plasma level in late controls (221 pg/mL) was lower than that in early controls (577 pg/mL) (P<.001). Women with early-onset preeclampsia had a decreased median plasma level of PlGF (27 pg/mL) compared with early controls (577 pg/mL) (P<.001). Women with late-onset preeclampsia had a lower median plasma PlGF level (48 pg/mL) than late controls (221 pg/mL) (P=.01). The median plasma level of PlGF in women with early-onset preeclampsia and a small for gestational age (SGA) infant (8.2 pg/mL) was lower than that in women with early-onset preeclampsia and an average for gestational age (AGA) infant (61 pg/mL) (P=.007).
Nonpregnant women had a median plasma level of VEGF-A of 35 pg/mL (Fig. 1C). In 67% of early controls, in 71% of late controls, in 69% of the women with early-onset preeclampsia, and in 65% of the women with late-onset preeclampsia, the plasma levels were under 15 pg/mL (the detection limit of the ELISA test), and therefore all pregnant groups had a median plasma level of 15 pg/mL. There were no significant differences between the study groups with respect to median plasma level of VEGF-A (P=.27) (Fig. 1C).
This study supports the suggestion that preeclampsia is associated with a disturbed relationship between pro- and antiangiogenic factors. This disturbance was found to be more pronounced in early- than in late-onset preeclampsia.
We found increased plasma levels of sFlt1 in both early- and late-onset preeclampsia compared with controls. Women with early preeclampsia had an approximately 43 times higher median level of sFlt1 compared with early controls. In late-onset preeclampsia, the corresponding value was three times higher than that in the controls. Our findings are consistent with data presented by Chaiworapongsa et al,20 who found a 10-fold increase in median plasma sFlt1 in early preeclampsia (diagnosed at 34 weeks of gestation or less) compared with early controls and a 2-fold increase in late preeclampsia (diagnosed at more than 37 weeks of gestation) compared with the controls.
In our study the mean length of gestation at delivery in women with early-onset preeclampsia was less than 29 weeks. The mean gestational age at parturition in the group of women with early-onset preeclampsia was not reported in the study by Chaiworapongsa et al.20 In our study the early-onset preeclampsia group may have been delivered more prematurely and may therefore have included more severe cases of preeclampsia, which would also lead to more pronounced changes in sFlt1. In addition, our results are consistent with those in a longitudinal study by Levine et al,13 who found elevated levels of sFlt1 5 weeks before the onset of preeclampsia and a larger change in the sFlt1 level in women with earlier onset of the disorder. The data were primarily cross-sectional and hence no further distinction into subgroups could be made.
In a study by Robinson et al22 including women with early-onset (delivered in gestational week 31–36, described as “severe PE”) and late-onset (delivered in gestational week 36–39, described as “mild PE”) preeclampsia, they could not find any difference in sFlt1-levels in women with early- compared with late-onset preeclampsia. Levine et al13 found serum levels of sFlt1 increasing during normal pregnancy, and we therefore consider that comparisons between different groups should be made in women at corresponding weeks of gestation.
In the present study, there was a difference in median levels of PlGF between early and late controls, with higher levels observed in early compared with late controls. This is consistent with previous longitudinal studies,9,17,23 where increasing levels have been found during normal pregnancy up to the end of the second trimester, followed by a decrease during the third trimester.
The relative decreases in the median plasma PlGF levels in the preeclampsia groups, compared with their respective control groups, were larger in early-onset preeclampsia (21-fold) than in late-onset preeclampsia (5-fold) (Fig. 1B). Women with early preeclampsia giving birth to SGA infants had a lower median PlGF level than those with early-onset preeclampsia giving birth to AGA infants. Our study thus indicates a progressively greater change in PlGF from late-onset preeclampsia to early-onset preeclampsia with AGA infants and, finally, the most pronounced changes in women with early-onset preeclampsia and SGA infants. Our findings are consistent with earlier reports of an association between the PlGF levels in women with preeclampsia and the levels in women giving birth to SGA infants, with the lowest levels in pregnancies complicated by both.17,24
In earlier studies extremely low concentrations of VEGF-A have been found during pregnancy.17,25 This is in accordance with our finding that 69% of the pregnant women (both women with preeclampsia and controls) had VEGF-A levels below the detection limit (15 pg/mL). We were, therefore, unable to investigate if there was a difference in VEGF-A concentrations between women with preeclampsia and controls. To address this question there is a need for more sensitive test systems.
There were some differences between women with preeclampsia and their controls regarding some baseline characteristics. The early control group had a lower body mass index in the first trimester compared with women who later developed early-onset preeclampsia. This was expected, because obesity is a risk factor for preeclampsia.26 Earlier studies have not shown any correlation between body mass index and sFlt1 or PlGF levels,13 and therefore the higher body mass index in women with early-onset preeclampsia in the present study should not influence our findings concerning sFlt1 or PlGF. The length of gestation at delivery was lower in women developing late-onset preeclampsia (38 weeks±2 standard deviation [SD]) than in their controls (40 weeks±1 SD). In normal pregnancy the sFlt1 level increases by 145 pg/mL per week after gestational weeks 33–36,13 and the levels of PlGF are supposed to decrease after 30 weeks of gestation.13 This means that we may have slightly underestimated the differences in plasma concentrations of sFlt1 and PlGF between women with late-onset preeclampsia and controls.
In contrast to most earlier studies of the association between angiogenic factors and preeclampsia, we have made a distinct separation between early- and late-onset disease. The mean length of gestation at delivery was 29 (±3 SD) weeks in women with early-onset preeclampsia and 38 (±2 SD) weeks in those with late-onset preeclampsia. In our population we noted differences between early- and late-onset preeclampsia regarding several fetal and maternal, clinical, and biochemical findings (Table 1). Early-, but not late-onset preeclampsia was associated with having a child born small for gestational age. At delivery, early-onset disease had a greater effect on the systolic blood pressure compared with late-onset disease, even though more women with early-onset preeclampsia were treated with antihypertensive medication. Compared with the controls, early, but not late preeclampsia had both an increased level of C-reactive protein and creatinine. These clinical findings suggest major clinical differences between early- and late-onset preeclampsia.
In the current study we have presented data on measurements at only one time point for each woman. As shown in previous studies,13–17 the alterations in sFlt and PlGF precede clinical signs of preeclampsia by several weeks. This finding has attracted attention because it raises a hope that tests for these parameters could be used as a screening procedure for preeclampsia.
In conclusion, we found differences in plasma levels of the proangiogenic factors VEGF-A and PlGF and antiangiogenic factor sFlt1 between early- and late-onset preeclampsia. In particular, we found a large difference in the plasma levels of sFlt1 between women with early-onset preeclampsia and their controls, compared with a moderate difference seen between women with late-onset disease and their controls. We also found a large difference in plasma levels of PlGF between women with early-onset preeclampsia and their controls, compared with the moderate difference seen between women with late-onset disease and their controls. These findings support the heterogeneity between early- and late-onset preeclampsia.
1. Roberts JM, Hubel CA. Is oxidative stress the link in the two-stage model of pre-eclampsia? Lancet 1999;354:788–9.
2. Redman CW, Sargent IL. Latest advances in understanding preeclampsia. Science 2005;308:1592–4.
3. Karumanchi SA, Bdolah Y. Hypoxia and sFlt-1 in preeclampsia: the “chicken-and-egg” question. Endocrinology 2004;145:4835–7.
4. Ness RB, Roberts JM. Heterogeneous causes constituting the single syndrome of preeclampsia: a hypothesis and its implications. Am J Obstet Gynecol 1996;175:1365–70.
5. Vatten LJ, Skjaerven R. Is pre-eclampsia more than one disease? BJOG 2004;111:298–302.
6. Odegard RA, Vatten LJ, Nilsen ST, Salvesen KA, Austgulen R. Preeclampsia and fetal growth. Obstet Gynecol 2000;96:950–5.
7. Aardema MW, Saro MC, Lander M, De Wolf BT, Oosterhof H, Aarnoudse JG. Second trimester Doppler ultrasound screening of the uterine arteries differentiates between subsequent normal and poor outcomes of hypertensive pregnancy: two different pathophysiological entities? Clin Sci (Lond) 2004;106:377–82.
8. Raijmakers MT, Dechend R, Poston L. Oxidative stress and preeclampsia: rationale for antioxidant clinical trials. Hypertension 2004;44:374–80.
9. Maynard SE, Min JY, Merchan J, Lim KH, Li J, Mondal S, et al. Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest 2003;111:649–58.
10. Ahmad S, Ahmed A. Elevated placental soluble vascular endothelial growth factor receptor-1 inhibits angiogenesis in preeclampsia. Circ Res 2004;95:884–91.
11. Stepan H, Faber R, Dornhofer N, Huppertz B, Robitzki A, Walther T. New insights into the biology of preeclampsia. Biol Reprod 2006;74:772–6.
12. Shibuya M. Structure and function of VEGF/VEGF-receptor system involved in angiogenesis. Cell Struct Funct 2001;26:25–35.
13. Levine RJ, Maynard SE, Qian C, Lim KH, England LJ, Yu KF, et al. Circulating angiogenic factors and the risk of preeclampsia. N Engl J Med 2004;350:672–83.
14. Tidwell SC, Ho HN, Chiu WH, Torry RJ, Torry DS. Low maternal serum levels of placenta growth factor as an antecedent of clinical preeclampsia. Am J Obstet Gynecol 2001;184:1267–72.
15. Levine RJ, Lam C, Qian C, Yu KF, Maynard SE, Sachs BP, et al. Soluble endoglin and other circulating antiangiogenic factors in preeclampsia. N Engl J Med 2006;355:992–1005.
16. Chaiworapongsa T, Romero R, Kim YM, Kim GJ, Kim MR, Espinoza J, et al. Plasma soluble vascular endothelial growth factor receptor-1 concentration is elevated prior to the clinical diagnosis of pre-eclampsia. J Matern Fetal Neonatal Med 2005;17:3–18.
17. Taylor RN, Grimwood J, Taylor RS, McMaster MT, Fisher SJ, North RA. Longitudinal serum concentrations of placental growth factor: evidence for abnormal placental angiogenesis in pathologic pregnancies. Am J Obstet Gynecol 2003;188:177–82.
18. Koga K, Osuga Y, Yoshino O, Hirota Y, Ruimeng X, Hirata T, et al. Elevated serum soluble vascular endothelial growth factor receptor 1 (sVEGFR-1) levels in women with preeclampsia. J Clin Endocrinol Metab 2003;88:2348–51.
19. Nadar SK, Karalis I, Al Yemeni E, Blann AD, Lip GY. Plasma markers of angiogenesis in pregnancy induced hypertension. Thromb Haemost 2005;94:1071–6.
20. Chaiworapongsa T, Romero R, Espinoza J, Bujold E, Mee Kim Y, Goncalves LF, et al. Evidence supporting a role for blockade of the vascular endothelial growth factor system in the pathophysiology of preeclampsia. Young Investigator Award. Am J Obstet Gynecol 2004;190:1541–7; discussion 1547–50.
21. Albertsson-Wikland K, Luo ZC, Niklasson A, Karlberg J. Swedish population-based longitudinal reference values from birth to 18 years of age for height, weight and head circumference. Acta Paediatr 2002;91:739–54.
22. Robinson CJ, Johnson DD, Chang EY, Armstrong DM, Wang W. Evaluation of placenta growth factor and soluble Fms-like tyrosine kinase 1 receptor levels in mild and severe preeclampsia. Am J Obstet Gynecol 2006;195:255–9.
23. Krauss T, Pauer HU, Augustin HG. Prospective analysis of placenta growth factor (PlGF) concentrations in the plasma of women with normal pregnancy and pregnancies complicated by preeclampsia. Hypertens Pregnancy 2004;23:101–11.
24. Lam C, Lim KH, Karumanchi SA. Circulating angiogenic factors in the pathogenesis and prediction of preeclampsia. Hypertension 2005;46:1077–85.
25. Polliotti BM, Fry AG, Saller DN, Mooney RA, Cox C, Miller RK. Second-trimester maternal serum placental growth factor and vascular endothelial growth factor for predicting severe, early-onset preeclampsia. Obstet Gynecol 2003;101:1266–74.
26. Doherty DA, Magann EF, Francis J, Morrison JC, Newnham JP. Pre-pregnancy body mass index and pregnancy outcomes. Int J Gynaecol Obstet. 2006;95:242–7.