The paper by Mulder et al. addresses the effect of pregnancy prolongation on maternal and fetal outcomes in women with early-onset pre-eclampsia diagnosed after 24 weeks of gestation (Mulder et al., BJOG 2020 xxxx). They report that pregnancy prolongation - from the time of pre-eclampsia diagnosis to delivery - is associated with improved offspring outcome and survival, without adverse consequences on short-term maternal cardiovascular and metabolic function. The maternal findings are apparently at odds with another recent study from New York (Rosenbloom et al. Obstet Gynecol 2020;135:27-35) which observed an increased risk of maternal cardiovascular events after pregnancy, in case of an interval of more than 7 days between the diagnosis of any hypertensive disorders of pregnancy and delivery.Some issues should be pointed out in order to avoid misunderstandings about these findings. Being Mulder et al.’s an observational study, readers cannot infer causality because women were not randomized to the length of pregnancy prolongation. Data are spread over a significant time period (from 1996 to 2017), and this aspect could be another confounder. Pre-eclampsia is a multi-organ syndrome based on chronic inflammation, oxidative stress and endothelial dysfunction leading to a persistent subclinical cardiovascular impairment and an increased risk of adverse events later in life (Sciatti et al., Eur J Prev Cardiol 2020 doi: 10.1177/2047487320925646), similarly to what happens in cases with heart failure with preserved ejection fraction. Myocardial geometry and ejection fraction are not sensitive enough to be altered by just a few days of pregnancy prolongation, and to forecast cardiovascular consequences. Only innovative techniques such as speckle-tracking imaging may document an impairment in myocardial contractility and relaxation in former pre-eclamptics, even if ejection fraction is normal.Pre-eclampsia is currently defined as new-onset hypertension combined with de-novo proteinuria and/or “adverse conditions” or “severe maternal/fetal complications” (Magee et al., Pregnancy Hypertens 2014;4:105-145). International guidelines recommend that women with severe forms of pre-eclampsia should be delivered immediately regardless of gestational age, while an expectant management should be considered for women with non-severe pre-eclampsia before term (Magee et al., Pregnancy Hypertens 2014;4:105-145; NICE guideline no. 133, 2019). Delaying delivery is expected to benefit newborn’s health, which is well exemplified by Mulder et al.’s findings. However, the fetus is often the protagonist of adverse conditions, and severe complications such as fetal growth restriction, often co-exist with early-onset pre-eclampsia, requiring longitudinal monitoring with Doppler ultrasound and cardiotocography. The timing of delivery depends on both maternal and fetal conditions. The lack of data about fetal growth and Doppler and cardiotocography findings (in cases with growth restricted babies) limits the generalisability of the results by Mulder et al. One would expect that a longer delay before delivery can be attained only in fetuses and mothers with milder clinical manifestations of disease.Word count: 455

The prenatal diagnosis of fetal anomalies started with the development of X-ray. In 1943, Hartley and Burnet, Radiologists in Manchester (J Obstet Gynaecol Brit Empire,1943;50:1-12), reported a series of 11 cases of “croaniolacunia” or lacunar skull, a condition often associated at births with spina bifida or encephalocele. These cases were all diagnosed in the third trimester of pregnancy and the radiograph features believed to be due to the effect of increased intracranial pressure on the fetal skull of hydrocephaly. Until the end of the 1960s, radiography remained the main technique to diagnose congenital abnormalities. In 1969, Russell (J Obstet Gynaecol Br Commonw,1969;76:345-50), also a consultant radiologist from Manchester, compared the accuracy of antenatal radiology examinations with paediatric reports in the diagnosis of anencephaly and other major neural tube defects, skeletal abnormalities such as achondroplasia and severe exomphalos when associated with rib deformities. Overall, the accuracy of the radiological diagnosis was considered as “strikingly” accurate for neural tube defects with 88 out of 113 cases of anencephaly diagnosed before delivery. Although, the author did not provide the gestational age at diagnosis, the images included in the article indicate that these were obtained in the third trimester. As neonatal care and surgery were in their infancy at the time, the main objective in diagnosing these anomalies was not the fetus but the need to identify antenatally mothers at risk of obstructed labour.Not surprisingly, some of the first publications by the team of Ian Donald in Glasgow were on the antenatal use of ultrasound imaging in the evaluation of the size of the fetal head (Willocks et al., J Obstet Gynaecol Br Commonw,1964;71:11-20). The fetal head was the only structure that could be measured and biparietal diameter the only measurement that could be obtained with the “ultrasound beam” of the first ultrasound machine (Figure). The technique called “cephalometry” was used at the end of the third trimester to assess “growth and maturity” of the fetus and “disproportion” and was found to be more reliable with ultrasound than X-ray. It would be another decade, before ultrasound imaging could reliably identify fetal anomalies such as spina-bifida in the second trimester of pregnancy (Campbell et al., Lancet,1975;1(7920):1336-7). However, the use of ultrasound imaging in the mid-seventies to search for major neural tube defects was always triggered by high levels of maternal serum alpha-fetoprotein. As there were few ultrasound equipment available and few trained operators, this biomarker was to remain for two decades the first line of action in the antenatal screening strategy for spina-bifida. The advent of high-resolution imaging, access low-cost and mobile ultrasound equipment and the training of more specialists and sonographers has moved the antenatal screening and diagnosis of many fetal anomalies to 11-14 weeks of gestation (Ushakov et al., UOG,2019;54:740-5).The systematic review by Drukker et al. (BJOG 2020) brings the focus back to late pregnancy: even in our exciting modern era of early anomaly scanning, a fetal abnormality will still be found in about 1 in 300 women scanned in the third trimester.Word count: 500BJOG since 1902 Perspectives on BJOG-20-0525R1