Hypertension in Pregnancy
Zwertbroek, Eva
DOI:
10.33612/diss.127418195
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from
it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date:
2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Zwertbroek, E. (2020). Hypertension in Pregnancy: Timing of delivery and early screening.
https://doi.org/10.33612/diss.127418195
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
General discussion and conclusions
Chapter 9
GENERAL DISCUSSION AND CONCLUSIONS
Part I Timing of delivery in late preterm hypertensive disorders Results in context
Obstetric studies generally evaluate short-term neonatal outcomes. However, the long-term consequences of obstetric management might be even more important. HYPITAT II was the first RCT on hypertensive disorders to conduct a long-term follow-up study at two time points, evaluating the effects of obstetric management on the offspring (chapters 2 and 3). In the 2-year follow up we found an increased rate in abnormal developmental outcome in the immediate delivery group as compared to expectant management. This finding strengthens previous studies reporting on the association between premature birth and an impaired developmental outcome.1,2 More precisely, the risk of developmental delay increases with decreasing gestational age at birth below 36 weeks.3 In addition, children in the 2- and 5- year follow up study had an increased rate of abnormal neurodevelopmental scores as compared to their peers from normotensive pregnancies in the Dutch population4. This is in line with previous findings that severe preeclampsia was associated with abnormal ASQ score in 28% of the children at the age of 2 compared to 19% in the control group.5,6 The timing of delivery mattered. In women with severe preeclampsia, children born preterm more often had abnormal ASQ scores (25%) at the age of 2 than children born at term(14%) although this was not statistically significant.7 The effect of hypertensive disorders and the timing of delivery on the offspring at the age of 5 has not been investigated before. The difference in development we found between the management groups at the age of 2 did not persist at the age of 5. At 5 years of age, environmental factors may become more important and may mitigate any persistent effects of pregnancy or delivery. Besides, the small difference in gestational age at delivery between the groups (6 days) is less likely to result in a difference in developmental outcome 5 years after birth.8 Although no difference between the management groups was found, children with abnormal outcomes at the age of 2 are at increased risk of abnormal outcomes at the age of 5. This long-term follow up study is new evidence regarding the effect of management in hypertensive disorders on the offspring at two timepoints in childhood. Children born after a hypertensive disorder in pregnancy are at increased risk of developmental problems in early childhood. Earlier delivery (on average 6 days) increases the risk. Identification of these children at risk of developmental problems is necessary to allow for early intervention.
The individual participant data meta-analysis in chapter 5 is currently the best available evidence regarding optimal timing of delivery of hypertensive disorders beyond 34 weeks of gestation. The currently available randomized controlled trails evaluated different outcomes, gestational ages, and hypertensive disorders. Individual results are presented in chapter 5.9-13 The pooled data resulted in a sample of 1724 women included in this IPD-MA. This allowed us to study rare non-composite outcomes, harmonize inclusion and exclusion criteria and to perform subgroup analyses. Our findings show that immediate delivery did significantly decrease HELLP and eclampsia (NNT 51), but it significantly increased RDS (NNH 58), but this depended on gestational age. This specific reduction of maternal morbidity has not been demonstrated before: the HYPITAT I study showed mainly a reduction of progression to severe disease (major part of the effect in the composite outcome) in the immediate delivery group.9 The HYPITAT II was not able to demonstrate a statistically significant reduction in severe maternal morbidity (including HELLP and eclampsia).10 RDS was shown to increase with immediate delivery in the late preterm period (34-37 weeks gestation). This was confirmed by the IPD-MA, but the effect depended on gestational age. At 35 weeks of gestation RDS was significantly increased in the immediate delivery group, whereas at 36 weeks and up the RDS risk was lower and did not significantly differ from the expectant monitoring group. At 37 weeks of gestation RDS risk diminished in both management groups. Recently published, new evidence regarding timing of delivery in women with preeclampsia was provided by the Phoenix trial. The results of 901 women showed that immediate delivery between 34 and 37 weeks of gestation significantly reduced maternal severe disease and complications, but it significantly increased neonatal admission rates.14 The authors clarified that the increase in admission was mainly due to prematurity without an increase in other neonatal morbidity such as RDS. In combination with our IPD-MA, this trial provides additional evidence for decision making in women with preeclampsia. The IPD-MA conducted in chapter 5 currently provides the best available evidence for timing of delivery regarding rare adverse outcomes that could influence clinical practice.
The challenge of rare outcomes
In our IPD-MA we used HELLP syndrome and eclampsia as the primary maternal outcome and RDS as the primary neonatal outcome. Outcome definition in hypertensive disorders of pregnancy is difficult. Serious adverse outcomes, which are important to prevent, are very rare. Surrogate and composite outcomes are often used, but the combination of a systolic blood pressure above 160 mmHg
with eclampsia/stroke, differ largely in severity and prevalence, making such a composite outcome both biologically and clinically hard to interpret. Which complications do we need to prevent, and at what cost? Is HELLP syndrome by itself severe enough to prevent at the cost of a preterm delivery between 34 and 37 weeks? Is preventing the development of severe hypertension enough to justify delivery after 36 weeks? How important are pediatric outcomes? For example, the HYPITAT I trial was criticized for the fact that the maternal composite outcome included severe hypertension (170/110 mmHg). In this study the composite outcome was indeed reduced by immediate delivery, but the effect was mainly due to the severe hypertension component. It was argued whether severe hypertension as a surrogate for adverse maternal morbidity justified delivery in ‘otherwise well women’. In the IPD-MA we were able to study rare solitary outcomes, not within a composite. The outcomes used were severe enough to prevent and easier for clinical interpretation.
Alternatively we could have included other severe complications of hypertensive disorders such as stroke, hepatic and renal failure, pulmonary edema and thromboembolic process. However, data regarding these outcomes were not available for IPD in all RCTs. An example of an adverse (rare) maternal core outcome set could be the fullPIERS outcomes including maternal death and adverse central nervous system outcomes, cardiorespiratory outcomes, hematological outcomes, hepatic outcomes, renal outcomes or placental abruption.14,15 Optimal outcome definition remains subject of debate.
Clinical implications
In this thesis we aimed to contribute to clinical decision making regarding optimal timing of delivery in hypertensive disorders of pregnancy. Current international guidelines vary in their recommendations. The American College of Obstetricians and Gynecologists (ACOG) suggests delivery at 37 weeks of gestation in the presence of gestational hypertension or pre-eclampsia.16 On the other hand, National Institute for Health and Care Excellence (NICE) UK guidelines recommends this only for women with pre-eclampsia.17 The HYPITAT I trial resulted in an increase in induction of labour rates in both women with gestational hypertension and preeclampsia in the Netherlands, as recommended by the NVOG (Dutch society of obstetrics and gynaecology).18 The IPD-MA described in this thesis was conducted to strengthen the evidence
on management of hypertensive disorders of pregnancy beyond 34 weeks of gestation and to potentially overcome heterogeneity in international guidelines. Based on our findings we recommend immediate delivery for all hypertensive disorders beyond 37 weeks of gestation. It reduces adverse maternal outcomes, not only severe disease, also in women with gestational hypertension. Beyond 37 weeks of gestation RDS rates are very low and not different in the two management groups (chapter 5). Moreover, long-term abnormal developmental outcomes in the offspring decrease significantly when delivered after 37 weeks of gestation.7
In the late preterm period, maternal outcomes favour delivery, whereas short-term neonatal outcomes favour expectant management. RDS risk increased when immediately delivered at 35 weeks and declined towards term. In light of our findings regarding unfavorable long-term pediatric outcomes (chapters 2 and 3) after immediate preterm delivery, clinicians should be reluctant to induce labor in the late preterm period. However, results of the IPD-MA showed that subgroups of women at increased risk of adverse maternal outcomes, such as women with preeclampsia, might benefit from earlier deliver. Immediate delivery in women with preeclampsia (with additional risk factors e.g. chest pain or dyspnea, low platelets, decreased oxygen saturation, progressively abnormal renal or liver enzymes tests) could be considered at 36 weeks, as RDS risk is relatively low. The results of the recently published Phoenix trial showed that in women with preeclampsia between 34 and 37 weeks of gestation immediate delivery decreases adverse maternal outcomes, whereas it increases neonatal admissions related to prematurity without increasing other neonatal morbidity.14 However, long-term pediatric outcomes of the Phoenix trial should be further evaluated before implementing this into routine clinical practice.
Future research
Future research should focus on long-term pediatric outcomes of obstetric interventions in hypertensive disorders of pregnancy. In fact, long term-follow up of the Phoenix trial is already planned. Differences found mainly in neonatal outcomes between our IPD-MA, the HYPITAT II study, and the Phoenix trial need further clarification. We are planning a new IPD-MA including only women with preeclampsia from randomized controlled trials. The currently available individual participant data of women with preeclampsia could be extended with the Phoenix data to provide better evidence. Such an extended IPD-MA
of women with preeclampsia would allow for more detailed analyses for this specific group of women with more advanced disease. Outcomes could be compared per week of gestation to identify optimal timing of delivery for women with preeclampsia in the late preterm period. Also, uncertainty remains about the effect of immediate delivery versus expectant monitoring on fetal growth. Expectant management might increase fetal growth restriction and the number of neonates being born small for gestational age. As low birthweight (<p10) has long-term health consequences, this outcome of immediate versus deferred delivery in hypertensive disorders remains to be investigated from the obtained IPD. Furthermore, evidence is lacking on neurodevelopmental outcomes per week of gestation at delivery. This is especially the case in the late preterm/ term period, between 36 and 38 weeks of gestation. Larger trials are needed to establish the effect of an obstetric intervention on long-term pediatric outcomes. This would guide clinicians further in decision making for immediate delivery or expectant monitoring.
Prediction of adverse outcomes Results in context
Although routine immediate delivery in the late preterm period is not justified, some women may benefit from earlier delivery. A prediction model could aid clinicians in predicting who is at high risk of adverse outcomes, to allow for early intervention. The model we developed in chapter 4 seems accurate and easy to use with clinical parameters: maternal age, gestational age at onset, systolic blood pressure, the presence of chronic hypertension, platelet count, creatinine and lactate dehydrogenase. The model could be used in all hypertensive disorders, in contrary to other models. Once externally validated, the prediction model developed in chapter 4 has the potential to be implemented in Dutch clinical care.
Various studies attempted to predict adverse outcomes in hypertensive disorders of pregnancy. Biomarkers are currently more frequently integrated in prediction models. It has been demonstrated that an increased ratio of proangiogenic factors (such als PlGF) to antiangiogenic factors (such as sFLT1 and soluble endoglin) is associated with a high risk of preeclampsia.19 If it is not increased, this ratio is especially useful as a criterion to rule out the occurrence of preeclampsia on the short term.20 Once preeclampsia is present, the sFlt1/PlGF ratio might predict adverse outcomes within 2 weeks.21 Based on these studies, the sFlt1/
PlGF test should be implemented and become routinely available in standard care. However, these biomarkers are not yet available in clinical practice. Other models that could be implemented in clinical practice are the fullPIERS and PREP models. They were both developed in large cohorts, include readily available clinical parameters, and show good performance. Moreover, both are externally validated in different settings and across countries, with still a good discriminative and predictive performance. The PREP model predicts complications in early-onset preeclampsia, including early preterm delivery within 48 hours and at discharge from the hospital.22 The model could be used to triage high-risk mothers who may need transfer to tertiary units for intensive maternal and neonatal care and timed delivery. The fullPIERS model predicts adverse outcomes (maternal death and adverse central nervous system, cardiorespiratory, hematological, hepatic and renal outcomes or placental abruption) within 48 hours, and up to 7 days after diagnosis.15,23 Variables included in the model are: gestational age, presence/absence of dyspnoea or chest pain, oxygen saturation, platelets, creatinine and AST/ALT. Performance was overall good depending on the external validation cohort. A risk above 30% is considered as the high risk cut-off for rule in of adverse outcomes up to 7 days after assessment. The fullPIERS model has also been validated in low-income settings.24
Clinical implications & future perspectives
The model we developed has to be externally validated before it can be implemented in clinical practice. An advantage above the international developed models is that our model could be used for all hypertensive disorders and is based on variables available in the perinatal health care system in the Netherlands. On the other hand, since the fullPIERS and PREP models are validated in many different populations, and are performing well, we would recommend them to be implemented in the Dutch clinical practice. These models could modify patient care by place and frequency of antenatal care. Low risk patients could be monitored in the outpatient department, and high risk patients could be monitored more frequently, admitted to the hospital and targeted for delivery. NICE guidelines recently advised to use the fullPIERS model in clinical practice. The 30% threshold that is used to identify women at high risk for inpatient management seems cost-effective in the late preterm period, but this should be further investigated.25 After implementation of the fullPIERS and PREP models in the Dutch guidelines, results and effect on obstetrical health
care should be evaluated. In addition, researchers should be encouraged to share data to be able to externally validate and update existing models. Further studies can focus on biomarker discovery and biomarker integration and implementation in routine care.
Figure 1. Flow chart of timing of delivery
* with onset after 34 weeks of gestation
Part II First trimester screening of preeclampsia Results in context
In the prospective pilot study (chapter 6), performance of the FMF first trimester prediction model was satisfactory for early and preterm preeclampsia and less sufficient for late preeclampsia. The algorithm performed poorly to stratify women at low risk and high risk. The small sample size and low outcome rate could be an explanation for this poor stratification. Moreover, it is known that the influence of the uterine artery Doppler measurement on the performance of the screening is substantial. In the cohort as described in chapter 6 (n=362 women), the MoM of the uterine artery Doppler was not associated with occurrence of preeclampsia. In the reproducibility study (chapter 7) of the uterine artery Doppler (n= 101 women), intra observer reproducibility was moderate (operator was experienced, 6 years after the FMF certificate for the uterine artery Doppler with over 1000 measurements). Inter observer reproducibility was moderate in case of experienced operators, whereas it was poor in case of a less experienced operator (6 months after the FMF certificate, 100 measurements). The full study cohort was measured by different sonographers, with different levels of experience. Therefore, the poor reproducibility of the measurement performed by less experienced sonographers could have largely influenced
the prediction model’s performance. With further training and experienced operators screening’s performance is likely to improve.
Detection rates in combination with false positive rates are inferior to the results reported by the group that developed the algorithm. 26,27 Tan et al. recently reported detection rate (DR) of 90% for early preeclampsia, 75% for preterm preeclampsia and 41% for term preeclampsia, at a false positive rate (FPR) of 10%.28 Based on the optimal cut-offs we identified in our validation study we found a DR for early and preterm preeclampsia, of 80% and 70%, respectively, but a much higher FPR of 19% for both. The DR for preeclampsia <42 weeks was 68%, at the cost of a FPR of 32%. Noteworthy is that, Tan et al. included over 61000 cases and the screening was performed by very experienced sonographers, using a highly standardized technique. Interestingly, cut-offs were similar in our cohort and the study of Tan et al., indicating similarity between our study populations. We assume that in a larger study cohort, with better standardization of the uterine artery Doppler measurement and more experienced sonographers performing all measurements, a similar performance would be achievable. Many first trimester prediction models have been developed over the past years.29 They all include a combination of maternal characteristics, ultrasound markers and biomarkers. Their clinical implication remains unclear, since very few have been externally validated. Besides the FMF preeclampsia screening algorithm, another first trimester prediction model with a substantial sample size has been externally validated.30 Scazzocchio et al., however, report good ROC curves but detection rates at fixed FPR of 10% are inferior to the FMF algorithm. In terms of clinical implementation, research should further focus on external validation of existing good performing prediction models across populations rather than on development of new prediction models.31
Apart from the technical aspects of developing and validating a screening model, the question whether screening would be relevant and beneficial is still under debate. Although there is no actual ‘cure’, preventive treatment options seem to delay the onset of preeclampsia. While health care professionals may think that screening would improve prenatal health care, an important factor with respect to screening uptake is the patient’s perspective. Opponents of preeclampsia screening speculate that patients may not be willing to undertake this ‘time consuming’ extra appointment at the hospital for an additional ultrasound screening and that taking another screening test would provide uncertainty
and anxiety. A study conducted in the Netherlands addressed these questions. Women were generally positive towards first trimester screening.32 Even when the test is not 100% accurate, 80% of the women still find the test informative. Reassurance, provided by a low-risk result, was the major motivation for undergoing the test. Self-monitoring, early recognition and intensive monitoring were considered benefits of using prediction models in case of a high-risk. Women acknowledged that high-risk determination could cause (unnecessary) anxiety, but it was expected that personal and professional interventions would level out this anxiety.32
Clinical implications
Based on accuracy of the FMF preeclampsia screening algorithm evaluated in chapter 6 and 7 preeclampsia screening in the Netherlands is feasible. However, performance should improve with further training of the sonographers, and standardization of measurements before successful implementation into clinical practice. Clinical implications of the screening algorithm require discussion of current clinical practice and challenges for implementation. Another ongoing discussion is if an extra screening is cost-effective with respect to the alternative of giving aspirin to all at risk women.
Current guidelines
Current standard of care regarding aspirin administration for prevention of preeclampsia as recommended by the National Institute for Health and Care Excellence (NICE, antenatal care) is as follows. The guideline identifies women at high risk for preeclampsia when they have any one high-risk factor (hypertensive disease in previous pregnancy, chronic hypertension, chronic renal disease, diabetes mellitus or autoimmune disease) or any two moderate-risk factors (nulliparity, age ≥ 40 years, body mass index (BMI) ≥ 35 kg/m2, family history of PE or interpregnancy interval > 10 years).17 The Dutch guideline is similar but adds BMI > 35kg/m to the high risk factors as well as twin pregnancy, placental insufficiency in previous pregnancy, and ovum donation to the moderate risk factors. The International Society for the Study of Hypertension in Pregnancy (ISSHP) and the American Congress of Obstetricians and Gynecologists (ACOG) use similar risk factors to NICE but their recommendations differ slightly.33,34 These guidelines are described in detail in chapter 6. The guidelines’ diagnostic accuracy was evaluated in two large studies.35,36 Using the high risk NICE guidelines, only 40% of the patients that will develop preterm preeclampsia were identified as high risk. Using the high risk definition for preeclampsia of
the American Congress of Obstetricians and Gynecologists (ACOG), 66% of the women were screen positive with detection rates of 89% and 90% for early-onset and preterm preeclampsia, respectively. However, the ACOG only recommends aspirin for women with a history of preeclampsia in at least two pregnancies or preeclampsia requiring delivery < 34weeks. This resulted in a detection rate of 6% of early preeclampsia and 5% for preterm at a false positive rate of 0.2%. In view of these poor detection rates, a better screening model is needed. Maternal factors provide important information for preeclampsia risk assessment. Including additional predictive markers that either reduce or increase the risk of preeclampsia would enhance the predictive accuracy and reduce the number of false-positive results.
Challenges of screening by maternal characteristics and biomarkers
The FMF screening model is externally validated and currently the best performing prediction algorithm.26,27,36,37 External validation studies report similar AUC’s but lower detection rates, as is also demonstrated in chapter 6.38 This could be due to variation in the a piori preeclampsia risk across populations
or measurement errors. Although performance in external validations studies seems accurate, there are some challenges to be faced prior to implementation. A large challenge remains the uterine artery Doppler measurement, especially in terms of implementation. Across the Netherlands but also in the UK, first trimester uterine artery Doppler measurement is not performed as part of routine care for the ‘general’ obstetrical population. These measurements are generally performed by sonographers working in second-third level centers. In order to implement this screening, adequate training of sonographers working in first level ultrasound practices should occur. This requires both a large financial and time investment. In addition, an extra visit to define the risk profile of the woman, measure the MAP, PLGF and uterine arteries PI should be scheduled. This is currently not yet standard practice in the Netherlands. These logistic challenges play against screening implementation. However, this might change with the implementation of a routine 13 weeks scan offered to all pregnant women. A concern for the performance of the screening remains the reproducibility of uterine artery Doppler measurements in the first trimester (chapter 7), unless a very stringent training of sonographers is performed. Alternatively, it may well be possible that in the near future automated measurement of the uterine artery may overcome the training and reproducibility issues. Another challenge of implementing the algorithm is that at present, with the disappearance of the combined test in the Netherlands, due to the introduction of NIPT as first
tier screening, first trimester maternal serum markers assessment should be introduced again.
Although the complete FMF algorithm (maternal risk factors, MAP, uterine artery PI, PlGF) achieves optimal performance (DR 75% of preterm PE cases), it may be decided to exclude a variable from the algorithm to adapt to the local setting. The combination of maternal risk factors, MAP and uterine artery PI detects 73% of PE cases and screening costs reduce by excluding PlGF. Screening by maternal risk factors, MAP and PlGF detects 68% of preterm PE cases.35 These alternative combinations of markers might increase feasibility of implementation. In any case, combinations of markers, added to the maternal characteristics perform much better than maternal risk factors alone.26,27
Aspirin for all?
Since the use of low-dose aspirin is safe during pregnancy, the question might arise: why don’t we administer aspirin to all pregnant women?39 Independently of the real safety of prolonged aspirin treatment given to all pregnant women, there are a few aspects to consider: compliance to treatment among all pregnant women would be lower if they are not aware of being at high-risk for preeclampsia. In addition, aspirin for all may give false reassurance and reduce awareness of high risk women. In terms of effectiveness, the reduction found in preterm preeclampsia in the ASPRE trial was based on a high risk population. Benefits might be much smaller or even nihil in a low risk population.40 Unnecessary treatment of millions of women might be a consequence, which could lead to increase in side effects.
Future perspectives
Future research should focus on the cost-effectiveness of preeclampsia screening and improving the quality of measurements of the parameters in the FMF algorithm. With respect to cost-effectiveness, a study in the USA demonstrated that aspirin for high risk women was more cost-beneficial than aspirin for all or no intervention at all.41 The idea is that screening could reduce cost of maternal and neonatal healthcare related to preeclampsia. However, further studies are needed to assess maternal and perinatal morbidity and mortality to assess the possible benefits of screening.
In addition, the FMF model also has its limitations; the predictive markers are not good enough to be used alone. Further studies could focus on new and strong
predictive markers (cell-free fetal DNA, proteomics, etc), or new combinations of existing markers. An IPD-MA for developing a new model and external validation would be an option. In addition, although preterm preeclampsia is associated with severe maternal and perinatal morbidity, term preeclampsia covers the largest burden of disease due to its high prevalence. Better methods to predict and prevent term preeclampsia are therefore needed.
The FMF screening algorithm has the potential to be implemented into clinical practice. It has its limitations, but is currently the best available option for preeclampsia screening. Screen positive rates and detection rates are superior as compared to current guidelines, even in small external validation cohorts. It is crucial to ensure that mean arterial blood pressure and the uterine artery pulsatility index are measured according to standardized techniques.42,43 More studies on the cost-effectiveness of preeclampsia screening and the impact on maternal and perinatal adverse outcomes are therefore needed.
REFERENCES
1. Petrini JR, Dias T, McCormick MC, Massolo ML, Green NS, Escobar GJ. Increased risk of adverse neurological development for late preterm infants. J Pediatr. 2009;154(2):169-176.
2. Johnson S, Evans TA, Draper ES, et al. Neurodevelopmental outcomes following late and moderate prematurity: A population-based cohort study. Arch Dis Child Fetal Neonatal Ed. 2015;100(4):F301-8.
3. Kerstjens JM, de Winter AF, Bocca-Tjeertes IF, Bos AF, Reijneveld SA. Risk of developmental delay increases exponentially as gestational age of preterm infants decreases: A cohort study at age 4 years. Dev Med Child Neurol. 2012;54(12):1096-1101.
4. Steenis LJ, Verhoeven M, Hessen DJ, van Baar AL. Parental and professional assessment of early child development: The ASQ-3 and the bayley-III-NL. Early Hum Dev. 2015;91(3):217-225.
5. Many A, Fattal A, Leitner Y, Kupferminc MJ, Harel S, Jaffa A. Neurodevelopmental and cognitive assessment of children born growth restricted to mothers with and without preeclampsia. Hypertens Pregnancy. 2003;22(1):25-29.
6. Tuovinen S, Raikkonen K, Kajantie E, et al. Hypertensive disorders in pregnancy and intellectual abilities in the offspring in young adulthood: The helsinki birth cohort study. Ann Med. 2012;44(4):394-403.
7. Warshafsky C, Pudwell J, Walker M, Wen SW, Smith GN, Preeclampsia New Emerging Team. Prospective assessment of neurodevelopment in children following a pregnancy complicated by severe pre-eclampsia. BMJ Open. 2016;6(7):e010884-2015-010884.
8. Mollborn S, Lawrence E, James-Hawkins L, Fomby P. When do socioeconomic resources matter most in early childhood? Adv Life Course Res. 2014;20:56-59.
9. Koopmans CM, Bijlenga D, Groen H, et al. Induction of labour versus expectant monitoring for gestational hypertension or mild pre-eclampsia after 36 weeks’ gestation (HYPITAT): A multicentre, open-label randomised controlled trial. Lancet. 2009;374(9694):979-988.
10. Broekhuijsen K, van Baaren GJ, van Pampus MG, et al. Immediate delivery versus expectant monitoring for hypertensive disorders of pregnancy between 34 and 37 weeks of gestation (HYPITAT-II): An open-label, randomised controlled trial. Lancet. 2015.
11. GRIT Study Group. A randomised trial of timed delivery for the compromised preterm fetus: Short term outcomes and bayesian interpretation. BJOG. 2003;110(1):27-32.
12. Boers KE, Vijgen SM, Bijlenga D, et al. Induction versus expectant monitoring for intrauterine growth restriction at term: Randomised equivalence trial (DIGITAT). BMJ. 2010;341:c7087. 13. Owens MY, Thigpen B, Parrish MR, et al. Management of preeclampsia when diagnosed between
34-37 weeks gestation: Deliver now or deliberate until 37 weeks? J Miss State Med Assoc. 2014;55(7):208-211.
14. Chappell LC, Brocklehurst P, Green ME, et al. Planned early delivery or expectant management for late preterm pre-eclampsia (PHOENIX): A randomised controlled trial. Lancet. 2019;394(10204):1181-1190.
15. von Dadelszen P, Payne B, Li J, et al. Prediction of adverse maternal outcomes in pre-eclampsia: Development and validation of the fullPIERS model. Lancet. 2011;377(9761):219-227.
16. American College of Obstetricians and Gynecologists, Task Force on Hypertension in Pregnancy. Hypertension in pregnancy. report of the american college of obstetricians and gynecologists’ task force on hypertension in pregnancy. Obstet Gynecol. 2013;122(5):1122-1131.
17. National Collaborating Centre for Women’s and Children’s Health (UK). Hypertension in pregnancy: The management of hypertensive disorders during pregnancy. . 2010.
18. de Sonnaville CMW, Hukkelhoven CW, Vlemmix F, et al. Impact of hypertension and preeclampsia intervention trial at near term-I (HYPITAT-I) on obstetric management and outcome in the netherlands. Ultrasound Obstet Gynecol. 2019.
19. Levine RJ, Lam C, Qian C, et al. Soluble endoglin and other circulating antiangiogenic factors in preeclampsia. N Engl J Med. 2006;355(10):992-1005.
20. Zeisler H, Llurba E, Chantraine F, et al. Predictive value of the sFlt-1:PlGF ratio in women with suspected preeclampsia. N Engl J Med. 2016;374(1):13-22.
21. Rana S, Powe CE, Salahuddin S, et al. Angiogenic factors and the risk of adverse outcomes in women with suspected preeclampsia. Circulation. 2012;125(7):911-919.
22. Thangaratinam S, Allotey J, Marlin N, et al. Prediction of complications in early-onset pre-eclampsia (PREP): Development and external multinational validation of prognostic models. BMC Med. 2017;15(1):68-017-0827-3.
23. Ukah UV, Payne B, Karjalainen H, et al. Temporal and external validation of the fullPIERS model for the prediction of adverse maternal outcomes in women with pre-eclampsia. Pregnancy Hypertens. 2019;15:42-50.
24. Ukah UV, Payne B, Lee T, Magee LA, von Dadelszen P, fullPIERS and miniPIERS Working Groups. External validation of the fullPIERS model for predicting adverse maternal outcomes in pregnancy hypertension in low- and middle-income countries. Hypertension. 2017;69(4):705-711.
25. National Institute for Health and Care Excellence. Hypertension in pregnancy: Evidence review for prediction of complications in pre-eclampsia.
26. NICE guideline NG133. https://www.nice.org.uk/guidance/ng133/evidence/c-prediction-of-complications-in-preeclampsia-pdf-6836186128. Updated 2019. Accessed 11/23, 2019.
27. O’Gorman N, Wright D, Poon LC, et al. Accuracy of competing-risks model in screening for pre-eclampsia by maternal factors and biomarkers at 11-13 weeks’ gestation. Ultrasound Obstet Gynecol. 2017;49(6):751-755.
28. O’Gorman N, Wright D, Syngelaki A, et al. Competing risks model in screening for preeclampsia by maternal factors and biomarkers at 11-13 weeks gestation. Am J Obstet Gynecol. 2016;214(1):103. e1-103.e12.
29. Tan MY, Syngelaki A, Poon LC, et al. Screening for pre-eclampsia by maternal factors and biomarkers at 11-13 weeks’ gestation. Ultrasound Obstet Gynecol. 2018;52(2):186-195.
30. De Kat AC, Hirst J, Woodward M, Kennedy S, Peters SA. Prediction models for preeclampsia: A systematic review. Pregnancy Hypertens. 2019;16:48-66.
31. Scazzocchio E, Crovetto F, Triunfo S, Gratacos E, Figueras F. Validation of a first-trimester screening model for pre-eclampsia in an unselected population. Ultrasound Obstet Gynecol. 2017;49(2):188-193. 32. Allen RE, Zamora J, Arroyo-Manzano D, et al. External validation of preexisting first trimester
preeclampsia prediction models. Eur J Obstet Gynecol Reprod Biol. 2017;217:119-125.
33. Crombag NM, Lamain-de Ruiter M, Kwee A, et al. Perspectives, preferences and needs regarding early prediction of preeclampsia in dutch pregnant women: A qualitative study. BMC Pregnancy Childbirth. 2017;17(1):12-016-1195-2.
34. Brown MA, Magee LA, Kenny LC, et al. Hypertensive disorders of pregnancy: ISSHP classification, diagnosis, and management recommendations for international practice. Hypertension. 2018;72(1):24-43.
35. American College of Obstetricians and Gynecologists, Task Force on Hypertension in Pregnancy. Hypertension in pregnancy. report of the american college of obstetricians and gynecologists’ task force on hypertension in pregnancy. Obstet Gynecol. 2013;122(5):1122-1131.
36. Tan MY, Wright D, Syngelaki A, et al. Comparison of diagnostic accuracy of early screening for pre-eclampsia by NICE guidelines and a method combining maternal factors and biomarkers: Results of SPREE. Ultrasound Obstet Gynecol. 2018;51(6):743-750.
37. O’Gorman N, Wright D, Poon LC, et al. Multicenter screening for pre-eclampsia by maternal factors and biomarkers at 11-13 weeks’ gestation: Comparison with NICE guidelines and ACOG recommendations. Ultrasound Obstet Gynecol. 2017;49(6):756-760.
38. Rolnik DL, Wright D, Poon LCY, et al. ASPRE trial: Performance of screening for preterm pre-eclampsia. Ultrasound Obstet Gynecol. 2017;50(4):492-495.
39. Park FJ, Leung CH, Poon LC, Williams PF, Rothwell SJ, Hyett JA. Clinical evaluation of a first trimester algorithm predicting the risk of hypertensive disease of pregnancy. Aust N Z J Obstet Gynaecol. 2013;53(6):532-539.
40. Mone F, Mulcahy C, McParland P, McAuliffe FM. Should we recommend universal aspirin for all pregnant women? Am J Obstet Gynecol. 2017;216(2):141.e1-141.e5.
41. Subtil D, Goeusse P, Puech F, et al. Aspirin (100 mg) used for prevention of pre-eclampsia in nulliparous women: The essai regional aspirine mere-enfant study (part 1). BJOG. 2003;110(5):475-484.
42. Werner EF, Hauspurg AK, Rouse DJ. A cost-benefit analysis of low-dose aspirin prophylaxis for the prevention of preeclampsia in the united states. Obstet Gynecol. 2015;126(6):1242-1250.
43. Poon LC, Zymeri NA, Zamprakou A, Syngelaki A, Nicolaides KH. Protocol for measurement of mean arterial pressure at 11-13 weeks’ gestation. Fetal Diagn Ther. 2012;31(1):42-48.
44. Plasencia W, Maiz N, Bonino S, Kaihura C, Nicolaides KH. Uterine artery doppler at 11 + 0 to 13 + 6 weeks in the prediction of pre-eclampsia. Ultrasound Obstet Gynecol. 2007;30(5):742-749.
