Fetal growth restriction (FGR) signifies that the fetus has not achieved its growth potential and is associated with increased perinatal mortality and morbidity. The exact aetiology of FGR, in the absence of any identifiable fetal and maternal factors, remains unclear and is attributed to placental insufficiency. The FGR placenta has a characteristic phenotype including: increased resistance in the fetoplacental circulation, an alteration in trophoblast cell turnover and reduced activity of placental nutrient transport systems, the best characterised being the amino acid transporter, system A. The placenta strongly expresses the cortisol inactivating enzyme, 11β-hydroxysteroid dehydrogenase type 2 (11β-HSD2). 11β-HSD2 activity was reduced in placentas from pregnancies complicated by FGR, suggesting increased exposure of the fetoplacental unit to maternal cortisol. In animal models, excessive exposure to glucocorticoids (GCs) is associated with a reduction in both fetal and placental weight. This reduction in placental weight was associated with abnormalities in placental function, consistent with those observed in the FGR placenta. This PhD investigated whether excess GC exposure during pregnancy is responsible placental insufficiency in human pregnancies and tested the hypotheses that excess GC exposure adversely affects placental vascular tone, trophoblast cell turnover and activity of the amino acid transporter, system A. Term placentas were collected from uncomplicated pregnancies and first trimester placental samples were obtained following elective surgical termination of pregnancy. Wire myography was used to explore the acute and chronic effects of GCs on term chorionic plate artery (CPA) function. The impact of GC treatment on trophoblast cell turnover in both first trimester and term placenta was investigated using the placental explant system. The effect of GCs on the activity of the system A transporter was also investigated in term explants and in isolated cytotrophoblasts where the expression of 11β-HSD2 was reduced using siRNA. Gene microarray studies on first trimester placental explants treated with GCs were utilised to identify genes regulated by GCs. Blunted constriction to thromboxane A2 was observed following acute GC treatment, whilst chronic exposure resulted in enhanced vasoconstriction, mimicking the altered reactivity of CPAs from pregnancies complicated by FGR. GC excess in first trimester placental explants increased apoptosis and decreased proliferation, thereby replicating the disordered turnover of the trophoblast observed in FGR placentas. No demonstrable effect was observed in cell turnover or system A activity in term placental explants treated with GCs, however, these experiments were hindered by the in-vitro regeneration of the syncytiotrophoblast in the model employed. The attenuation of 11β-HSD2 activity observed in FGR placentas was replicated in term primary cytotrophoblasts utilising siRNA to knock-down expression of 11β-HSD2. Preliminary results suggested an increase in system A activity in response to cortisol. Gene microarray studies identified a significant number of genes (~500) that were regulated by dexamethasone, confirming that GCs have an impact on many aspects of placental function. Potential mediators for the characteristic features of the FGR placenta replicated here in response to GC treatment were identified and validated at the mRNA level. The studies described in this thesis support the hypotheses that GC excess within the placenta contributes to the development of raised vascular resistance in the fetoplacental circulation and the disordered trophoblast turnover in placentas from pregnancies complicated by FGR. However, with the preliminary studies performed, the hypothesis that elevated levels of GCs contribute to the reduced placental amino acid transfer by the system A transporter in the FGR placenta can not be confidently disproven.

Fetal growth restriction (FGR), when a fetus fails to reach its genetic growth potential, affects up to 10 % of pregnancies and is a major risk factor for both neonatal and adulthood morbidity and mortality. There are currently no treatments for FGR except for delivery of the fetus; resulting in premature delivery which, in itself, is linked to poor outcome. Therefore, the focus of current research is to examine whether therapies successfully used to treat diseases with similar aetiologies to FGR can also be used to treat FGR. Sildenafil citrate (SC), a selective phosphodiesterase-5 inhibitor, is one such candidate. With the recent announcement of the STRIDER international clinical trial for the treatment of severe FGR with SC, it is imperative to determine the efficacy and safety of SC treatment on both fetus in utero and long-term adult health. Mouse models that mimic characteristics of human FGR represent an attractive model to perform pre-clinical studies. Recent studies in mice have demonstrated that SC increased fetal and placental weight and normalised umbilical artery blood flow velocity in FGR but no studies have assessed effects of antenatal SC on offspring health. The aims of this study were to assess the effect of antenatal SC treatment on a) fetal weight b) fetal vascular reactivity b) pup viability and d) long-term effects on postnatal development/physiology in a mouse model of FGR.All experiments were performed in the placental-specific insulin-like growth factor 2 knockout mouse (Igf2 P0+/- mice) which have mixed litters of wild-type (WT) and growth restricted (P0) mice. It has been reported that SC administered in the drinking water was able to increase P0 fetal weight and thus this mouse model was chosen to assess the effects of SC on the fetus and offspring. SC was administered to pregnant dams in two regimens; orally (120 – 160 mg.kg-1) and subcutaneously (10 mg.kg-1) between E12.5 and E18.5. WT and P0 fetal abdominal aortas were isolated at E18.5 and ex vivo vascular function was assessed using wire myography. Fetal abdominal aortas demonstrated reliable and reproducible vasocontraction and vasorelaxation; there were some sex- and genotype-specific differences. SC demonstrated dose-dependent effects on fetal aortic function. Offspring from dams treated with a subcutaneous injection of SC or saline were assessed for postnatal growth (week 5 – week 12), systolic blood pressure (week 8 and week 13), glucose tolerance (week 12) and mesenteric / aortic vascular function (week 14 – week 16). These experiments demonstrated that;• A supratherapeutic concentration of antenatal SC (120 – 160 mg.kg-1) did not increase fetal weight but significantly blunted relaxation responses of fetal abdominal aortas at E18.5. • A subcutaneous injection of antenatal SC (10 mg.kg-1) did not increase fetal weight or alter fetal abdominal aortic function in mice but led to increased systolic blood pressure in both WT and P0 offspring. Additionally, glucose sensitivity was significantly reduced in female offspring from SC treated dams. In conclusion, the studies outlined in this thesis have demonstrated that antenatal SC treatment can cause alterations in fetal blood vessel function and also lead to changes in metabolic and cardiovascular function in mouse offspring. Using ex vivo wire myography, mouse fetal abdominal aortas were able to be assessed at E18.5. This methodological advance will be beneficial as it can be applied to assessing putative treatments in mice that show characteristics of human FGR. In addition, this technique will allow for investigation of the underlying mechanisms of in utero programming of adulthood cardiovascular diseases such as hypertension. Future work must focus on the mechanisms leading to increased systolic blood pressure in offspring from SC treated dams and whether such effects are noted in other animal models of FGR using a variety of SC dosing regimens. These studies will provide information with which to increase efficacy, and ensure the safety, of SC treatment in pregnancy complications.

Pregnancy complications such as fetal growth restriction (FGR) are often attributed to poor uteroplacental blood flow, but the risk of systemic side-effects hinders therapeutic intervention. We have utilised novel placental-specific homing peptides to overcome this and have conjugated these to biocompatible liposomes. Peptide-conjugated liposomes were found to selectively bind to the outer syncytiotrophoblast layer of the human placenta and to the uteroplacental vasculature and labyrinth region of the mouse placenta. The novel vasodilator SE175 was selected as a nitric oxide donor with a favourable stability and release profile, to encapsulate in peptide-conjugated liposomes in an attempt to restore impaired uteroplacental blood flow in a mouse model of FGR, the endothelial nitric oxide synthase knockout mouse. Liposomes containing SE175 or PBS were prepared by lipid film hydration and targeting peptides coupled to the liposomal surface. Vehicle control, free SE175, PBS- or SE175-containing liposomes were intravenously injected on embryonic (E) days 11.5, 13.5, 15.5 and 17.5. Animals were sacrificed at E18.5 and fetal and placental weights recorded. Targeted delivery of SE175 significantly increased fetal weight compared to vehicle control but no other treatment groups, whilst significantly decreasing placental weight, indicating improved placental efficiency. Treatment was well tolerated, having no impact on litter size or resorptions. Targeted delivery of SE175, but no other treatment group, reduced a marker of lipid peroxidation in the placenta, indicating a reduction in oxidative stress. These data suggest that selective delivery of SE175 to the uteroplacental vasculature in peptide decorated liposomes may represent a novel treatment for FGR.

Adequate placental function is essential for the growth and development of a healthy fetus. A major cause of abnormal placental function is thought to occur from inadequate maternal spiral artery remodelling, leading to maternal vascular malperfusion (MVM) of the placenta and ultimately fetal growth restriction (FGR) and stillbirth due to uteroplacental hypoxia. Current methods of investigating a pregnancy at risk of FGR rely on ultrasound estimations of fetal size and Doppler studies. A more informative measure may be to quantify placental function in-vivo. Magnetic resonance imaging (MRI) has the ability to assess placental oxygen saturation (sO2), using Blood Oxygen-Level Dependent (BOLD), and the partial pressure of oxygen (pO2) using Oxygen-Enhanced MRI (OE MRI). These MRI techniques have been shown to correlate with gestation and poor pregnancy outcomes in cross sectional studies. MRI measures of placental oxygenation are hypothesised to be a potential antenatal tool for the identification and stratification of high risk pregnancies at risk of FGR related to uteroplacental hypoxia. To address this hypothesis changes in placental oxygenation, following maternal hyperoxia, were calculated in normal and FGR pregnancies in a cross sectional study. The change in placental oxygenation was reproduced longitudinally to determine if the rate of change differed between normal and FGR pregnancies. Baseline placental MRI parameters (R1 and R2*) and measures of the change in oxygenation were incorporated into a diagnostic model to identify FGR related to uteroplacental hypoxia, which was provisionally tested in a group of high risk pregnancies to demonstrate its potential clinical utility. Placental measures of baseline R1 and R2* were significantly increased in FGR pregnancies. The change in placental pO2 following hyperoxia was found to be significantly lower in FGR pregnancies. The change in pO2 declined similarly with gestation in both cross sectional and longitudinal studies, in normal and FGR pregnancies. There were no significant correlations in the change in placental sO2 with gestation or pregnancy outcome. The use of a diagnostic model combining baseline R1 and R2* and pO2 measures identified FGR with a high specificity, and provided additional information to aid in disease stratification and decision making in a significant proportion of the high risk pregnancies tested. In conclusion, MRI parameters of placental pO2 following hyperoxia are significantly lower in FGR pregnancies, in keeping with the concept of uteroplacental hypoxia. MRI techniques show promise in the identification of FGR pregnancies related to MVM through measures of placental function, irrespective of fetal size, and may aid in the disease stratification of high risk pregnancies.

Objetivos. Avaliar os fatores prognÃsticos para o Ãbito neonatal em gestaÃÃes com diÃstole zero ou reversa na dopplervelocimetria da artÃria umbilical MÃtodos. Estudo transversal a partir dos prontuÃrios das gestantes com diagnÃstico de diÃstole zero (DZ) ou reversa (DR) em artÃria umbilical acompanhadas no ServiÃo de Medicina Materno Fetal da Maternidade-Escola Assis Chateaubriand â Universidade Federal do CearÃ. Foram analisadas 48 pacientes com gestaÃÃo Ãnica, sem anomalias estruturais ou cromossÃmicas, apresentando idade gestacional superior a 22 semanas e menor do que 34 semanas. Para a avaliaÃÃo estatÃstica, foram empregados os testes: t Student, Exato de Fisher, Qui-quadrado de Pearson e RegressÃo LogÃstica e Multinomial. Todos foram considerados estatisticamente significantes quando p<0,05. Resultados. As sÃndromes hipertensivas foram observadas na maioria (78,3 %) das gestantes. A maioria (52,1%) das gestaÃÃes foi resolvida nas primeiras 24 horas apÃs o diagnÃstico dopplervelocimÃtrico. Em uma semana, 81,3% dos casos tinham terminado em parto, preferencialmente (85,4%) pela via abdominal. Os maiores percentuais da idade gestacional no momento do diagnÃstico dopplervelocimÃtrico concentraram-se na faixa de 25 e 27 semanas. Os pesos do RN variaram entre 550g e 2600g, com mÃdia de 1021,7g. Os recÃm-nascidos foram classificados como pequenos para idade gestacional em 79,1% dos casos. Ao primeiro minuto de vida, 24 (57,1%) RN apresentaram Ãndices de Apgar menores do que 7. Ocorreram 26 Ãbitos neonatais. Do estudo estatÃstico univariado dos fatores de risco antenatais, a idade gestacional no momento do diagnÃstico dopplervelocimÃtrico revelou-se variÃvel significativamente relacionada com o Ãbito neonatal (RR; 2,1, 95% CI 1.152 â 4.008, p = 0.011). Do estudo estatÃstico univariado dos fatores de risco pÃs-natais peso do RN (RR; 2,6, 95% CI 1.329 - 5.238, p = 0.001) e Apgar ao primeiro minuto (RR; 1,9, 95% CI 1.03 â 3.588,p = 0.027) revelaram-se variÃveis significativamente relacionadas com o Ãbito neonatal, com peso do RN apresentando sensibilidade de 76,9%, especificidade de 73,6%, valor preditivo positivo de 80,0% e valor preditivo negativo de 30%, e Apgar ao primeiro minuto apresentando sensibilidade de 61,5%, especificidade de 61,1%, valor preditivo positivo de 69,5% e valor preditivo negativo de 47,6%. ConclusÃes. DiÃstole zero ou reversa estÃo relacionadas com resultados perinatais adversos, cujo risco para Ãbito neonatal està relacionado com a idade gestacional no momento do diagnÃstico dopplervelocimÃtrico e com o peso do RN e com Apgar ao primeiro minuto. / Objectives: evaluate the perinatal results in pregnancies with fetal brain sparing on the Doppler velocimetric study and identify the main prognostic factors associated with neonatal death. Methods: it is a transverse study from the charts of pregnant wowen with diagnosis of brain sparing, absent or reversed end-diastolic flow in the umbilical artery, followed at the Service of Maternal-Fetal Medicine of Maternidade-Escola Assis Chateaubrind â Universidade Federal do CearÃ. There were analyzed 143 patients with single pregnancies, without structural or chromosomal anomalies, presenting gestacional age above 22 weeks and fetal weight equal or above 500 grams. ROC curve was constructed for gestacional age and weight at birth (independent variables) and neonatal death (dependent variable). The perinatal results were evaluated on the general population and on each group (brain sparing, absent and reversed end-diastolic flow), later compared with each other. For the statistical analisys it was utilized the tests: Shapiro-Wilk, Levene, t Studente, Mann-Whitney, ANOVA, Kruskal Wallis, Fisher. Chi-square, Logistical and Multinomial Regression. All were considered statistically significant when p < 0.05. Results: the majority of pregnante wowen (78.3%) presented some hypertensive disturb associated to the pregnancy. The pregnancy was resolved in the first 24 hours after Doppler velocimetric diagnosis on most cases (74.8%), being the abdominal acess utilized in 96.5% of the times. At the moment of delivers, the average gestational age was 33.6 weeks and the weight was 1684g. The newborns were classified as small for gestational age in 69.6% and needed ICU admission in 63% of the cases. The indexes of perinatal mortality for brain sparing, absent and reversed end-diastolic flow were respectively 11.1, 31.1 and 70.6%. The weight of the newborn (area bellow the ROC curve 0.934, p=0.000 and gestational age at birth (area 0.909, p=0.000) have shown to be good predictors of neonatal death. The cutoff point calculated for the weight was 1010g and for the gestational age as 32.5 weeks. The incidence of diminished amniotic fluid indez (AFI) in the pregnancies with lethal perinatal outcome was 41.2% and in those without lethality was 41.3%. Conclusions: fetuses with diagnosis of brain sparing, absent and reversed end-diastolic flow presented progressively worse and statistically different with each other prognosis. The gestational age and weight at birth showed excellent correlation with neonatal mortality. The AFI did not demonstrate association with lethality rate.

During normal placental development, a primitive vascular network develops through vasculogenesis and angiogenesis, and is then remodelled through maturation and regression. The mechanism behind this regression is unknown, but data from other systems suggests that it could be due to an endothelial-mesenchymal transition (EndMT). If this is the case, then dysregulated EndMT could lead to increased vascular regression, which could result in placental hypovascularisation. As the placental vasculature is the area of exchange between maternal and fetal circulations, a reduction in its surface area could result in fetal growth restriction (FGR). The hypothesis of this thesis is that EndMT occurs during normal placental development, but is increased during FGR and contributes to placental hypovascularisation. A primary cell model consisting of endothelial and mesenchymal cells was isolated from human first trimester placental villous stroma. These cells were shown to lose CD31 mRNA (n = 1-3) and protein (n = 15) over 4 passages, with no loss of cell viability (n = 8). EndMT-associated transcription factors were also present in these cells at all 4 passages (n = 2-4). When cells were isolated from this mixed cell model based on their CD31-positivity and examined immediately after isolation, a small proportion also expressed αSMA (n = 5). Co-expression of endothelial and mesenchymal markers suggests that an EndMT was occurring. After 24 hours in culture, the proportion of these cells expressing αSMA increased (n = 5), and some cells co-expressed vWF and αSMA, while others lost their CD31-positivity, indicating that these cells had undergone EndMT. Cells isolated based on their CD31-positivity were treated with factors shown to inhibit EndMT in other systems. However, culture with 10µM SB431542 (TGFβ receptor inhibitor; n = 6), 10µM Dorsomorphin (BMP receptor inhibitor; n = 3), or 0.1µM PDGFR-β Tyrosine Kinase Inhibitor IV (n = 3) did not inhibit gain of αSMA by these cells. Culture on Matrigel in endothelial growth medium containing VEGF and FGF also failed to stabilise the endothelial phenotype (n = 3). The possibility that EndMT occurs in placenta in vivo was examined; genes associated with EndMT were shown to be present in placenta (n = 5), and there was limited evidence of CD31 or vWF co-expression with αSMA in tissue. Preliminary evidence was obtained to suggest that expression of EndMT-associated genes was altered in FGR placentas compared to normal. In summary, the data presented in this thesis demonstrate that an EndMT occurs in primary placental microvascular endothelial cells in vitro. Furthermore, these studies provide evidence to suggest that this transition also occurs in vivo and could be altered in placentas from pregnancies complicated by FGR.

Advanced maternal age (AMA) is defined as childbearing in mothers ≥35 years of age and is becoming increasingly prevalent in high income countries. AMA has been associated with increased risk of adverse pregnancy outcomes, particularly stillbirth. Although AMA mothers have higher rates of chromosomal abnormalities and maternal co-morbidities, AMA remains an independent risk factor for stillbirth. Despite these findings, the etiology behind this increased risk is unknown. We hypothesise that an altered maternal environment, including increased oxidative stress and inflammation, due to ageing causes placental dysfunction which increases AMA mothers’ vulnerability to stillbirth. A holistic approach was applied to investigate placental dysfunction in AMA. Firstly, a systematic review and meta-analysis comprehensively reviewed existing data on AMA and associated adverse pregnancy outcomes. Secondly, Manchester Advanced Maternal Age Study (MAMAS), a multi-centre prospective observational cohort study, was conducted to investigate risk factors for composite adverse pregnancy outcome (CAPO) in AMA. MAMAS utilised both uni- and multivariate analysis on demographic and clinical data, and measuring biomarkers of ageing and placental dysfunction by ELISA in maternal circulation during the third trimester of pregnancy. Utero-placental dysfunction was directly investigated in uncomplicated AMA pregnancies by quantifying placental morphology, placental nutrient transport capabilities and both placental and maternal uterine vascular responses. Finally, a C57BL/6J murine model of AMA was developed and characterised to further investigate maternal age on pregnancy outcome and the role of the placenta. In the meta-analysis, maternal age was linearly associated with increased risk of stillbirth and other adverse outcomes strongly associated with placental dysfunction (fetal growth restriction, preeclampsia and placental abruption). In MAMAS, smoking status and primiparity were predictive of CAPO. After adjustment, AMA mothers had an odd ratio of 2.05-3.43 of CAPO compared to 20-30 year old mothers. AMA mothers showed evidence of increased oxidative stress and pro-inflammatory bias. AMA mothers who suffered CAPO showed reduced placental endocrine capacity seen in placental dysfunction. Placentas from uneventful AMA pregnancies showed evidence of accelerated ageing and placental adaptation with increased nutrient transport, increased placental weight but reduced efficiency, and altered vascular function. AMA mice showed many similar aspects to human AMA with increased fetal loss, fetal growth restriction and increased placental size. These studies provide robust evidence for increased incidence of adverse pregnancy outcome due to placental dysfunction in pregnancies of women of AMA. This finding requires the appropriate recognition in a clinical context, with a greater focus on personalised obstetric care in an attempt to reduce stillbirth rates in this high risk population. By optimising antenatal and obstetric care for AMA mothers, we could reduce stillbirth rates by 4.7% - the population attributable risk due to AMA. These studies highlight key areas of future research that will further understanding into stillbirth risk in AMA pregnancy, test predictive models and test therapies and clinical care interventions an ultimately improve pregnancy outcome in mothers of AMA.

No description available.

Abstract Abnormal placentation is associated with preeclampsia and placental insufficiency, both of which increase the risk for fetal growth restriction. So far the early recognition of the risk population for preeclampsia has been problematic. The first hypothesis of this study was that in preeclampsia, the maternal serum proteomic profile is different from that in uncomplicated pregnancies, and this difference is detectable already in early pregnancy. The findings of this study demonstrate that in clinical preeclampsia the maternal serum proteomic profile is different from that in uncomplicated pregnancies with increased levels of placental proteins and antiangiogenic factors in pregnancies with clinical preeclampsia. Furthermore, the early pregnancy maternal serum proteomic profile in women who later develop preeclampsia revealed a distinct and different pattern compared with the profile in clinical preeclampsia. In early pregnancy, the differentially expressed proteins belong to placental proteins, vascular and/or transport proteins and matrix and/or acute phase proteins, while angiogenic and antiangiogenic proteins were not significantly expressed in early pregnancy. Preeclampsia, placental insufficiency, fetal growth restriction and type-1 diabetes may have an impact on fetal cardiovascular hemodynamics. The second hypothesis in this thesis was that in placental insufficiency, abnormalities in fetal cardiovascular status correlate with biochemical markers of cardiac dysfunction and chronic hypoxia. In placental insufficiency, increases in fetal N-terminal pro-atrial (NT-proANP) and pro-B-type natriuretic peptide (NT-proBNP) and in fetal erythropoietin concentrations were related to increased pulsatility in the fetal umbilical artery and descending aorta. In addition, these fetuses demonstrated increased pulsatility in their systemic venous blood velocity waveforms. Thus, in placental insufficiency, biochemical markers of cardiac dysfunction and chronic hypoxia are associated with signs of increased fetal cardiac afterload and systemic venous pressure. Increased NT-proANP and NT-proBNP levels were also detected in fetuses of type-1 diabetic mothers with normal umbilical artery velocimetry. In these pregnancies, NT-proANP and NT-proBNP levels were related to poor maternal glycemic control during early pregnancy.

Abstract The first aim of this study was to investigate the relationship between Doppler ultrasonographic parameters and biochemical markers of human fetal cardiac dysfunction and myocardial cell damage in pregnancies complicated by placental insufficiency and/or fetal growth restriction. Our second aim was to examine fetal central and peripheral hemodynamic characteristics associated with retrograde aortic isthmus net blood flow. Fetuses with significant myocardial cell damage (cTnT > 0.10 ng/ml) had increased pulsatility in the blood velocity waveforms of ductus venosus, left hepatic vein and inferior vena cava, and had more often atrial pulsations in the umbilical vein. Their umbilical artery NT-proANP concentrations were higher than in fetuses without myocardial cell damage. The proportion of left ventricular cardiac output of the combined cardiac output was greater and the corresponding proportion of the right ventricle was less than in fetuses with only increased NT-proANP levels ( > 1145 pmol/l). Tricuspid regurgitation was present more often and the right ventricular fractional shortening was less in fetuses with myocardial cell damage than in fetuses with normal umbilical artery cTnT levels. In fetuses with placental insufficiency and/or growth restriction (n = 48), umbilical artery NT-proANP concentrations showed a significant positive correlation with ductus venosus, left hepatic vein and inferior vena cava pulsatility index values for veins. Fetuses with placental insufficiency and antegrade aortic isthmus net blood flow demonstrated a shift in their right ventricular cardiac output from the pulmonary to the systemic circulation, and foramen ovale volume blood flow made up the majority of the left ventricular cardiac output. Fetuses with retrograde aortic isthmus net blood flow failed to demonstrate these changes, and they had signs of increased left atrial pressure. In addition, right ventricular fractional shortening was decreased and the pulsatility in the ductus venosus blood velocity waveforms was increased. In conclusion, human fetal myocardial cell damage was associated with a rise in systemic venous pressure, a change in the distribution of cardiac output towards the left ventricle and a rise in right ventricular afterload. Fetuses with retrograde aortic isthmus net blood flow failed to rearrange the distribution of the cardiac output and they had signs of increased left atrial pressure. In addition, right ventricular afterload and pulsatility in the ductus venosus blood velocity waveforms were increased.