Association of biochemical markers with pre-eclampsia

Pre-eclampsia is a placenta-driven pregnancy complication causing systemic inflammation and endothelial dysfunction. It is characterised by increased blood pressure and proteinuria and affects vital organs of the body. Therefore, without surveillance and treatment it threatens the health and life of the mother and the foetus. Every year approximately 70,000 mothers die due to pre-eclampsia and its complications globally. There is still no cure for pre-eclampsia other than delivery. The aim of this thesis was to investigate associations of biomarkers to pre-eclampsia, and to find better ways to predict the disease and its subtypes, and accordingly to also increase the understanding of the mechanisms and pathways of the pathophysiology of pre-eclampsia, which eventually may lead to innovations of new therapeutic targets. This thesis includes data from the multidisciplinary PREDO (Prediction and prevention of pre-eclampsia and fetal growth restriction) project, which consists of 988 pregnant women with and 117 without known risk factors for pre-eclampsia. These women were recruited between September 2005 and December 2009 in ten maternity clinics in Finland. The usefulness of serum hyperglycosylated human chorionic gonadotropin (hCG-h) or the proportion of hCG-h to hCG (%hCG-h) was assessed for the prediction of pre-eclampsia in the first trimester (Study I). These were combined with maternal risk factors, known biomarkers (placental growth factor (PlGF), free human chorionic gonadotropin beta (hCGβ) and pregnancy-associated plasma protein A (PAPP-A)), and biophysical measurements (mean arterial pressure and mean uterine artery pulsatility index). The study was conducted in a subcohort of 257 women with known risk factors for pre-eclampsia. We found that prior pre-eclampsia and low serum %hCG-h were associated with late-onset and non-severe pre-eclampsia, whereas low serum PlGF was associated with early-onset and severe pre-eclampsia. Free hCGβ was higher in women who developed severe pre-eclampsia than in women who did not develop severe pre-eclampsia, whereas low serum PAPP-A was associated with non-severe pre-eclampsia. Multivariate models constructed with regularised logistic regression provided only modest prediction rates for all pre-eclampsia (36% sensitivity with 90% specificity) and its subtypes (20% sensitivity for early-onset pre-eclampsia and 32% sensitivity for late-onset pre-eclampsia with 90% specificity). The effect of low-dose aspirin (100mg/d) on maternal serum concentrations of PlGF during pregnancy from the first trimester to the late second trimester was investigated (Study II). Blood samples were collected at 12–14, 18–20 and 26–28 weeks of gestation. The main finding was that high-risk women who had low-dose aspirin treatment (N=61) started before 14 weeks of gestation for prevention of pre-eclampsia had higher concentrations of serum PlGF than high-risk women who had placebo treatment (N=62). The difference was evident from mid-gestation onwards. We investigated the differences of haemoglobin scavenger proteins haemopexin (Hpx) and alfa-1-microglobulin (A1M) in high risk women and low risk women at 26–28 weeks of gestation (Study III). It was a case–control study with high-risk women who subsequently developed pre-eclampsia (n=42), high-risk women who did not develop pre-eclampsia (n=49) and low-risk women (n=51). We found higher plasma Hpx concentration in high risk women, who developed pre-eclampsia compared to low risk women. Plasma A1M was higher in the group of high-risk women, who did not develop pre-eclampsia than in other groups. We investigated the changes in maternal plasma Hpx and A1M concentrations from the first trimester to late second trimester (blood sampling at 12–14, 18–20 and 26–28 weeks of gestation) (Study IV). We used the same study cohort than in the third study. It appeared that high-risk women, who did not develop pre-eclampsia had a unique profile of haemoglobin scavenger proteins during pregnancy. Firstly, unlike in other groups, plasma Hpx concentration did not change during the study period, and it was lower during the first half of the pregnancy than in other groups. Secondly, the A1M concentration increased during the first half of the pregnancy and stayed at the higher level compared to the other two study groups. During the first half of the pregnancy the change in A1M concentration in high-risk women who did not develop pre-eclampsia was opposite to the change seen in those women who developed pre-eclampsia, while there was no change of plasma A1M concentration in low-risk women. We also found that women who subsequently developed pre-eclampsia and gave birth to a small-for-gestational-age newborn had consistently higher plasma levels of A1M than women who developed pre-eclampsia and gave birth to an appropriate-for-gestational-age newborn. The difference was significant from mid-gestation onwards. In conclusion, the serum PlGF concentration of women who started low dose aspirin before 14 weeks of gestation was higher from mid-gestation onwards compared to the other groups. This may provide one mechanism by which low-dose aspirin prevents pre-eclampsia. This thesis sheds light on the serum haemoglobin scavenger protein dynamics during the first and second trimester of pregnancy in women who have low or high risk for pre-eclampsia, as it was found that higher plasma concentrations of A1M in high-risk women, who do not develop pre-eclampsia may be associated with a reduced risk of developing pre-eclampsia. Furthermore, high-risk women who do not develop pre-eclampsia may have unique, protective dynamics of serum haemoglobin scavenger proteins. These findings suggest that not only clarification of the pathophysiology of pre-eclampsia, but also protective factors should be a focus of future research. Additionally, it was shown that the change in plasma A1M concentration in women who subsequently develop pre-eclampsia may be inversely related with foetal growth from mid-gestation onwards until the late second trimester. Multivariate models constructed with regularised logistic regression provided only modest prediction rates for all pre-eclampsia and its subtypes. This thesis strengthens the theory that pre-eclampsia is a complex multi-factorial syndrome, in which many biochemical pathways are affected.