Search

The aim of my thesis was to examine the use of placental magnetic resonance imaging in preeclampsia and chronic hypertension. Placental magnetic resonance imaging techniques are currently in development and there is a paucity of data for their use and application in hypertensive disorders of pregnancy. The objectives of my research were to firstly optimise placental magnetic resonance imaging and associated measures in women with preeclampsia and chronic hypertension and to describe potential unexpected findings in magnetic resonance imaging performed for research during pregnancy. Secondly, I set out to provide an approach to visual assessment of the placenta in uncomplicated pregnancies using T2-weighted imaging. Thirdly I investigated placental changes in preeclampsia for an insight into the pathophysiology of preeclampsia and then investigated placental changes in chronic hypertension for an insight into the heterogeneity of pregnancy outcomes. Lastly, I explored changes in associated placental biomarkers Placental Growth Factor (PlGF), soluble fms-like tyrosine kinase 1 (sFlt-1), vascular adhesion molecule (VCAM) and hyaluronan in women with preeclampsia and chronic hypertension and described them in the context of placental T2* values obtained from magnetic resonance imaging. These objectives were achieved through performing a prospective, observational study, embedded within the National Institutes of Health (NIH) funded Placenta imaging Project, which aimed to develop a novel magnetic resonance approach to assess growth and development of the human placenta in health and disease. The focus of my project related specifically to evaluating placental magnetic resonance imaging in women with chronic hypertension and preeclampsia. In this thesis, the optimisation and application of advanced magnetic resonance imaging techniques have provided detailed placental imaging in women with preeclampsia and chronic hypertension. These techniques have enabled qualitative visual assessment and quantitative assessment of placental structure and function. Optimised T2-weighted imaging sequences have led to the development of a comprehensive approach to visual assessment of the placenta in uncomplicated pregnancies and are subsequently used as a reference for assessing women with preeclampsia and chronic hypertension. Quantitative assessment of the placenta has included the development of T2* mapping and diffusion sequences, culminating in an optimised combined diffusion-relaxometry sequence which provides regionally matched diffusion and T2* values in a reasonably fast and acceptable scan time compared to conventional sequences. Methods assessing placental T2* maps to quantify the visual variation seen are explored and detect variability within such maps. These measures include lacunarity as well as histogram derived measures of kurtosis and skewness, in addition to mean T2*. These advanced sequences of T2-weighted imaging, T2* mapping and diffusion weighted imaging have been applied to women with preeclampsia and chronic hypertension in order to explore the spectrum of placental phenotypes in these hypertensive disorders of pregnancy. In pregnancies complicated by preeclampsia, T2-weighted imaging showed substantial areas of low signal intensity, advanced lobularity and high granularity within lobules with a reduced entire placental mean T2* for gestational age and higher lacunarity values compared to uncomplicated pregnancies. In pregnancies complicated by chronic hypertension, T2-weighted imaging showed a varied visual appearance compared to gestation matched controls with some showing features similar to those with preeclampsia and some indistinguishable from those in the control group. Not all women with mean T2* values outside of the normal range developed adverse pregnancy outcomes and conversely not all women with normal mean T2* values had uncomplicated pregnancies. This suggest a more complex interaction between the placenta and maternal or fetal response, while the timing of imaging (in relation to delivery) may be crucial. Finally, T2* mapping was explored in conjunction with placental biomarkers of PlGF, sFlt-1, hyaluronan and VCAM to further elucidate mechanisms underlying preeclampsia and chronic hypertension. These biomarkers have been found to provide complementary mechanistic information of placental phenotypes seen with T2* mapping.

The aims of my PhD were to understand physiological and pathophysiological factors that influence outcomes in reproductive health and pregnancy for women with chronic kidney disease (CKD) before, during and after pregnancy. Standard assessment of renal function in pregnancy is by measurement of serum creatinine concentrations yet normal gestational ranges had not been established. Serum anti-Müllerian hormone (AMH) is a biomarker of ovarian reserve, but the clinical interpretation of AMH in women with CKD is ambiguous and had not been examined in early stage CKD. There are limited contemporary data available to inform counselling and surveillance of women with moderate and severe kidney disease (CKD stages 3-5) undertaking pregnancy, with outcomes restricted to small and historical cohorts. Mechanistic links between superimposed pre-eclampsia and CKD include endothelial dysfunction, renin-angiotensin system activation, complement activation and tubular injury, which offer the potential for novel diagnostic indicators. Placental growth factor (PlGF) concentrations in isolation and in combination with soluble fms-like tyrosine kinase-1 (sFlt-1:PlGF) have been recently implemented as diagnostic adjuncts in women with suspected pre-eclampsia, yet data on the utility of these markers in women with CKD are limited. Given that the diagnosis of superimposed pre-eclampsia in women with CKD is complicated by the presence of hypertension and proteinuria due to kidney disease, an evaluation of the utility of PlGF, sFlt-1, and other novel biomarkers in the prediction of superimposed pre-eclampsia in women with CKD was warranted. Methods used in this thesis included a systematic review and meta-analysis of 4421 serum creatinine concentrations in pregnancy, a prospective cohort study of AMH concentrations in 163 reproductive-age women with CKD, a retrospective cohort study of obstetric and renal outcomes in 178 pregnancies in women with pre-pregnancy CKD stages 3-5, a nested case-control study of novel biomarkers in the diagnosis of superimposed pre-eclampsia in women with CKD; and a prospective multicentre study of 232 pregnancies in women with CKD, examining the accuracy of placental growth factor (PlGF) and soluble fms-like tyrosine kinase-1 (sFlt-1) in predicting the need for delivery due superimposed pre-eclampsia. The upper limits of the reference interval for serum creatinine in pregnancy were 85%, 80% and 86% of the upper limit outside of pregnancy in sequential trimesters. This means that for an upper reference limit of 90µmol/L for serum creatinine in non-pregnant females, values greater than 76µmol/L in the first trimester, 72µmol/L in the second trimester, and 77µmol/L in the third trimester should be considered abnormal in pregnancy, and warrant investigation to exclude a diagnosis of CKD or acute kidney injury in pregnancy. Serum AMH concentrations in women with CKD aged less than 35 years were lower than in women without CKD, across all CKD stages. Women with CKD aged 20-24 years had comparable serum AMH concentrations to women aged 35 years and over without CKD. Pregnancies in women with pre-pregnancy CKD stages 3-5 were complicated by preterm delivery, low birthweight and loss of maternal renal function. Chronic hypertension was the strongest predictor of delivery before 34 weeks' gestation, with additional risk if the gestational fall in serum creatinine was less than 10% of pre-pregnancy values. Pre- or early pregnancy proteinuria was the strongest predictor of birthweight below the 10th centile. There was a step-decline in renal function in relation to pregnancy in most women with pre-pregnancy CKD stages 3-5, equivalent to between 1.7 and 4.9 years of background renal disease depending on pre-pregnancy CKD stage and rate of decline in kidney function prior to pregnancy. There was no evidence that renal transplantation conferred additional risk in women with pre-pregnancy CKD stages 3-5. Superimposed pre-eclampsia affected one third of women with CKD. Although plasma PlGF concentrations were lower in women with CKD who developed superimposed pre-eclampsia, mean concentrations did not fall below 100pg/ml. Plasma PlGF (Quidel) concentrations below 150pg/ml had the highest sensitivity and negative predictive value for the prediction of delivery due to superimposed pre-eclampsia in women with CKD. High plasma hyaluronan and VCAM concentrations discriminated both pre-eclampsia and superimposed pre-eclampsia from uncomplicated pregnancy, supporting existing pathogenic theories that pre-eclampsia is a disease of endothelial dysfunction. However, predictive performances for hyaluronan and VCAM were lower than for plasma PlGF concentrations. Quantification of PlGF, sFlt-1 and sFlt-1:PlGF ratio in serum did not usefully predict the need for delivery due to superimposed pre-eclampsia in women with CKD. There was no demonstrable diagnostic role for factors derived from the renin-angiotensin and complement systems. This thesis was a multifaceted study of a heterogeneous disease. It addressed knowledge gaps for women with CKD across the spectrum of reproductive health including disease definition, pre-pregnancy assessment, diagnosis and prediction of superimposed pre-eclampsia in pregnancy, and long-term renal outcomes. It also formed the basis for the first national guideline for women with kidney disease in pregnancy in the UK.

Introduction Intrahepatic cholestasis of pregnancy (ICP) is associated with adverse perinatal outcomes and maternal metabolic dysfunction. Ursodeoxycholic acid (UDCA) treatment can improve itch severity and reduce bile acids in cholestatic conditions. This thesis determines how UDCA affects ICP patients by assessing its impact on adverse perinatal outcomes, its role in reducing maternal pruritus, and its involvement in maternal metabolic dysfunction secondary to altered enteroendocrine signalling. Methods To understand the magnitude of adverse perinatal outcomes in women with ICP, I performed a systematic review and aggregate data meta-analysis, comparing women with ICP (5515) and uncomplicated pregnancies (165 081), using a random-effects comparison with the DerSimonian and Laird method. To determine whether these outcomes are affected by disease severity, I analysed individual patient data from 16 published studies and two unpublished cohorts and compared results for women untreated and treated with UDCA. To identify pruritogens that are modifiable by UDCA treatment, I analysed serum samples collected longitudinally from women with ICP and benign pruritus gravidarum. Results were compared with linear and logistic regression, and predictive biomarkers for ICP were identified. To determine the effect of ICP on enteroendocrine signals, I studied murine models (0.5% cholic acid-fed ± UDCA) and samples from women with ICP, uncomplicated pregnancy, and non-pregnant controls. I used 16S rRNA sequencing and whole genome shotgun metagenomic analyses to characterise the cholestatic gut microbiota, and assessed the gut metabolome using UPLC-MS. Their effects on enteroendocrine signals were determined using qPCR and enzymatic analyses on ileal tissues and serum samples obtained following a timed dietary study. Results ICP is associated with increased adverse perinatal outcomes, compared with uncomplicated pregnancies. Women with bile acids ≥100μmol/L have significantly 24 increased risks of stillbirth (hazard ratio 30.50 (8.83 to 105.30, p < 0.0001)) compared to women with lower total bile acids. UDCA treatment did not significantly affect the observed association between total bile acid concentration and stillbirth, although the observed cohort treated with UDCA had more severe disease than untreated women, and the absolute risk of stillbirth was reduced for treated women at the equivalent disease severity (determined by bile acid concentration). Sulfated progesterone metabolites and autotaxin activity are associated with disease severity (determined by pruritus intensity) for women with ICP compared to those with benign pruritis gravidarum. They are also useful predictive biomarkers prior to hypercholanaemia developing; and their levels are reduced by treatment with UDCA. UDCA treatment of women with ICP and a murine model of cholestatic pregnancy results in enrichment of the gut microbiota with Bacteroidetes compared to Firmicutes, with associated increased bile salt hydrolase activity and intestinal bile acid deconjugation. Unconjugated intestinal UDCA is thereby modified to the metabolically-active bile acid lithocholic acid, whose levels are markedly increased in the faeces of treated women. Discussion These studies have confirmed that ICP is associated with adverse perinatal outcomes, including stillbirth; UDCA treatment may lower the stillbirth risk at an equivalent bile acid concentration, but the existing evidence does not yet support this definitive conclusion. UDCA can improve the pruritus of ICP by reducing the pruritogenic sulfated progesterone metabolites, and may improve the metabolic derangements of ICP by altering the gut microbial environment to enhance bile acid-derived enterohepatic and enteroendocrine signalling.

Background: In 2015, over 800 women died in pregnancy and childbirth every day. There are effective treatments for the leading causes of maternal death, but they require early detection by measurement of vital signs, and timely administration to save lives. The CRADLE-Vital Sign Alert accurately measures blood pressure and pulse and calculates shock index. Results are displayed on a traffic light early warning system. This aim of this thesis was to evaluate the impact of this intervention on maternal mortality and morbidity in low and middle-income countries. Methods: A pragmatic, stepped-wedge randomised-controlled trial with a nested mixed-methods process evaluation was undertaken. This was preceded by a mixed-method feasibility study. The intervention was introduced into every level of routine maternity care across 10 clusters in Africa, India and Haiti. The primary composite outcome was at least one of eclampsia, emergency hysterectomy and maternal death per 10,000 deliveries. Delivery of the intervention, its uptake and potential mechanism of action were measured and integrated with qualitative findings and measures of resources and staffing in each cluster. Findings: Between April 1st 2016 and November 30th 2017, among 536,223 deliveries, the primary outcome occurred in 4067 women. There was an 8% decrease in the primary outcome from 79.4/10,000 deliveries pre-intervention to 72.8/10,000 post-intervention. After planned adjustments for variation in event rates between and within clusters over time, the unexpected degree of variability meant we were unable to judge the benefit or harms of the intervention (OR 1.22, 95% CI 0.73-2.06a p=0.45). Overall, the intervention was delivered with reasonable fidelity and improved the availability of vital signs equipment and number of women with BP measurements (79.2%-s. 97.6%a OR 1.30, 95% CI 1.29 - 1.31). Therewere significant differences in the effect of the intervention between 8 individual clusters which could not be explained by the measures of implementation or local context. Interpretation: The acceptability and feasibility of the intervention has been demonstrated. Despite the rigorous trial design, effectiveness of the intervention at reducing mortality and morbidity has not been shown. Measuring implementation alongside effectiveness was feasible and beneficial in describing differences between clusters in maternal health. In this case, these differences could not explain the difference in the effect of the intervention between individual clusters. Stepped-wedge trials across multiple countries ha -e considerable methodological challenges and should be powered to demonstrate an effect of the intervention in each country. Further research in the selection and integration of process measures in low and middle -income countries is required.

This thesis examines maternal and perinatal outcomes in pregnancy complicated by chronic hypertension. National guidance highlights the paucity of data from randomised controlled trials to guide choice of antihypertensive treatment for chronic hypertension in pregnancy. I performed a systematic review/meta-analysis of all randomised controlled trials comparing antihypertensive agents to non-active treatment/other antihypertensive agents for treatment of chronic hypertension in pregnancy. I demonstrated that antihypertensive agent use was associated with substantial reduction in the incidence of severe hypertension, but no significant differences in other maternal/perinatal outcomes were observed. The pathophysiology underpinning the increased risk of adverse maternal/perinatal outcome in women with chronic hypertension is poorly understood and likely to be multifactorial. I undertook a cohort study of 4481 pregnancies in women with chronic hypertension and assessed the impact of maternal characteristics on adverse perinatal outcome. Black ethnicity was strongly associated with adverse perinatal outcome. I conducted a randomised controlled feasibility study to compare labetalol and nifedipine for treatment of chronic hypertension in pregnancy. Feasibility of recruitment was confirmed and effectiveness of labetalol and nifedipine to control mean blood pressure to target examined. Outside pregnancy, Black women are recommended calcium-channel blockers as first-line antihypertensive treatment. I observed differences in treatment effect between ethnic groups and highlighted areas requiring further investigation. A nested mechanistic study within the trial demonstrated differences in placental/renal biomarker concentrations in women who developed superimposed pre-eclampsia, compared to those who did not, and between women of Black ethnicity compared to non-Black. Additional assessment of maternal vascular function across gestation showed no differences by ethnicity, but longitudinal variation in brachial blood pressure, central aortic pressure, pulse wave velocity, and augmentation index were demonstrated in women with chronic hypertension who developed adverse pregnancy outcome (superimposed pre-eclampsia and in those delivering small for gestational age infants) compared to those who did not.

Pre-eclampsia is a disease unique to pregnancy. Prevalence in the UK is between 5- 8% of pregnancies yet diagnosis remains challenging. The PELICAN study was a multi-centre, observational cohort study. The primary aim was to evaluate the diagnostic accuracy of plasma placental growth factor (PlGF) in the second half of pregnancy, in predicting the need for delivery for pre-eclampsia within 14 days of testing. 649 women presenting with suspected pre-eclampsia were recruited between January 2011 and February 2012, across seven consultant-led units within England and Ireland. Blood samples were taken at enrolment; PlGF measurements were performed but results blinded until the study was complete and diagnoses and pregnancy outcome known. A further 47 biomarkers were measured (using 57 assays) to evaluate whether the diagnostic potential of PlGF could be improved further. Using a pre-specified cut off of <5th centile, a low (>12pg/ml < 5th centile) or very low (<12pg/ml) PlGF concentration was shown to have high sensitivity (0.95 CI (0.89- 0.99) in women < 35 weeks’ gestation) to determine need for delivery within 14 days. When compared with other biologically plausible biomarkers, the area under the ROC curve for low or very low PlGF (0.87, standard error 0.03), was greater than all other commonly utilised tests either singly or in combination (range 0.58–0.76; p<0.001 for all comparisons). Data from 100 women were then used to perform a budget impact analysis. A hypothetical decision analytical model using data extracted from case note review and reference cost tariffs, suggested a mean cost saving associated with the PlGF test (in the PlGF plus management arm) of £35,087 (95% CI -£33,181 to -£36,992) per 1,000 women, equating to a saving of £582 (95% CI -£552 to -£613) per woman tested. PlGF testing could be used to risk-stratify women with suspected pre-eclampsia with the aim of improving pregnancy outcome.

Current techniques to identify growth-restricted fetuses, at risk of health complications, have limited accuracy. Placental insufficiency is a key pathological process in fetal growth restriction (FGR). I investigated the potential clinical benefit of placental biomarkers to identify pregnancies delivering small for gestational age (SGA) infants in pregnancies with suspected pre-eclampsia and in those with reduced symphysis-fundal height measurement using delivery of an SGA infant as a surrogate measure of FGR. Suspected pre-eclampsia (PELICAN-PE study) In a large multicentre prospective cohort study investigating diagnostic accuracy of placental growth factor (PlGF) in women with suspected pre-eclampsia, I assessed test performance of 47 biomarkers and ultrasound parameters to identify women delivering an SGA infant. PlGF measurement outperformed all other biomarkers and currently used tests in predicting delivery of an SGA infant. Combinations of biomarkers added minimal value. Reduced symphysis-fundal height measurement (PELICAN-FGR study) I assessed the ability of PlGF and ultrasound parameters to predict delivery of an SGA infant in women with reduced symphysis-fundal height (current UK standard to identify pregnancies at risk of SGA) in a second multinational prospective cohort study. Test performance statistics were calculated for all parameters in isolation and combination. Ultrasound parameters had modest test performance for predicting delivery of an SGA infant. PlGF performed no better. Incorporating PlGF with ultrasound parameters provided modest improvements. In women presenting with suspected pre-eclampsia, PlGF measurement is a potentially useful adjunct to current practice in identifying those at risk of SGA. The findings of the PELICAN-FGR study cannot support the use of PlGF to risk stratify women referred with reduced symphysis-fundal height. The prevalence of FGR in the two studies differed, with a high number of normal pregnancies in those presenting with reduced symphysis-fundal height. The pathological process in normotensive versus hypertensive SGA may differ, potentially explaining these findings.