Although antenatal care was widely introduced in the Western world only in the 20th century, it is now universally accepted and advocated as beneficial and advisable for all pregnant women and hence consumes a considerable amount of accommodation, equipment, professional time and costs to the women themselves. Resource use is rarely measured routinely but must vary enormously because of differences in the amount and intensity of care proposed, and in its uptake, in different geographical settings. However, even when resource use is at the lower end of the spectrum, antenatal care should be audited so that the needs of women and their babies can be best addressed.

Cerebral palsy has been defined as a group of non-progressive disorders of movement or posture due to a defect or lesion of the immature brain.1 Cerebral palsy is defined by clinical description only; there is no definitive test. It is not informative about aetiolgy, pathology or prognosis. The definition of cerebral palsy also lacks precision in that neither the age at which the brain is mature nor the severity of the disorder required are specified. Brain development is most rapid in utero and early childhood but continues at a diminished rate throughout life. There is therefore no biological rationale for choosing any particular age to define brain maturity. The defining age below whcih cerebral palsy may be recognised therefore varies between centres for pragmatic reasons usually associated with assertainment. For example, in Western Australia the criterion is set at 5 years,2 since by this age any central motor impairments of congenital or neonatal origin will be apparent and can usually be differentiated from progressive disorders or isolated motor delay. Thus the definition includes a group of children with post-neonatally acquired brain damage resulting in movement disorder. Post-neonatally acquired cerebral palsy differs from presumed congenital or perinatal cerebral palsy in that there is usually no doubt that it was caused by a specific post-neonatal event such as meningitis or head trauma. Although there is an excess of perinatal morbidity amoung children acquiring cerebral palsy post-neonatally,3 suggesting a prior vulnerability in some cases, post neonatally acquired cerebral palsy will not be considred further in this review.

Selective fetal growth restriction (sFGR) has been reported to occur in about 10–15% of monochorionic (MC) twins. The diagnosis of sFGR has been based on variable criteria including estimated fetal weight (EFW), abdominal circumference and/or the degree of fetal weight discordance. Recent studies tend to use a simple definition which includes the presence of an EFW less than the 10th percentile in the smaller twin. Some would argue that the intertwin fetal weight discordance should be included in the definition. Indeed this factor plays a major role in the complications presented by these cases. While the majority of cases with one fetus below the 10th percentile usually will also present with a large intertwin EFW discordance, the contrary is not always true. Thus, it is possible to find MC twins with remarkable intertwin EFW discordance but the EFW of both fetuses are still within normal ranges. Although it appears to be common sense that a large intertwin discrepancy might represent a higher risk for some of the complications described later in this review, there is no consistent evidence to support this notion. Therefore, due to its simplicity, a definition based on an EFW below 10th percentile in one twin is probably the most useful for clinical and research purposes.

Although amniocentesis was first reported in 1882, the technique was not in widespread clinical practice until the 1970's. The reason for this slow uptake was that there were few indications for performing the procedure until fetal karyotyping from amniotic fluid cells became possible in 1966. Currently fetal karyotyping is the commonest indication for the technique and amniocentesis has become the mainstay of antenatal diagnosis.

The techniques employed in the diagnosis of placenta praevia have altered radically since Guillemeau first described the condition in 1685, stating that ‘The surgeon must consider if it is the child or if it is the after-birth who presents first’. However, his description went unnoted, for Giffart in narrating a case of haemorrhage wrote in 1730: ‘I cannot receive as absolutely true the opinion of those authors, who say that the placenta is always attached to the fundus uteri, for in this case, as in many others, I have every reason to believe that it adhered on the internal orifice, or very near to it; and that, in dilating, the latter occasioned the separation of the after-birth, and as a consequence the haemorrhage.’

Meconium ileus (MI) is one of the most common causes of intestinal obstruction in the newborn accounting for 9–33% of neonatal intestinal obstructions. It is the earliest clinical manifestation of cystic fibrosis (CF), occurring in approximately 16% of patients with CF. However, MI has been reported in the absence of CF.

The use and method of intrapartum obstetric analgesia has remained controversial since the late 16th Century. In 1591, during the reign of James, VI, Eufame MacCalzean (a Scottish midwife) was the first to attempt pain relief during labour and childbirth. At that time, the provision of pain relief during labour was an unacceptable medical practice due mainly to religious and social objections. Because of these objections, and historical accounts that claimed the midwife had consulted with a witch during the delivery, both MacCalzean and the midwife were found guilty of witchcraft and subsequently executed.

Diagnostic ultrasound has an enviable reputation for safety, and the lack of evidence for risk has been one of the key factors which has established it as the pre-eminent imaging method in obstetrics. Ultrasound remains one of very few common diagnostic procedures which may be carried out with little regard to safety. By comparison, all radiographic examinations carry known genetic and somatic risks associated with ionising radiation. Magnetic resonance imaging (MRI) investigations have resulted in several hundred cases of superficial burns, trauma from projectiles, and deaths from malfunctioning pacemakers or displaced metallic inserts. Furthermore, these other modes give rise to biological effects in organs other than those being imaged. For example, a radioisotope lung scan exposes the uterus due to radioactive urine, and scattered x-radiation exposes the ovaries during a chest computed tomography (CT) examination. By comparison, all bio-effects from ultrasound are restricted to the scanned region of tissue.

Human and ovine fetuses produce and absorb prodigious amounts of water and electrolytes daily. As will be described, fetal fluid exchange approximates to 300–400ml/kg per day near term. On a per kilogram basis, this is equivalent to an adult ingesting and secreting 15–20 litres per day. Upon delivery, the newborn rapidly adjusts to a much reduced rate of fluid exchange. In this respect, the fetus resembles an amphibian during development: an aquatic phase of in utero gestation and a comparatively arid state following delivery. Although the amphibian-like pattern of development may be partly an example of ‘ontogeny recapitulates phylogeny’, the remarkable volume of fetal fluid exchange suggests an important physiological role for amniotic fluid.

For an infant without lethal congenital malformations there is no risk greater than to be born too early. In addition, preterm birth with its many consequences may compromise the life of the whole family. Thus, prevention of preterm birth is one of the greatest challenges in obstetrics. However, this has proven to be difficult. This difficulty is in part due to the fact that, although we know a large number of clinical factors which are know a large number of clinical factors which are associated with preterm birth, the final mechanisms triggering the onset of preterm contractions or premature rupture of the fetal membranes (PROM) have remained largely unclear. We review the prevention of preterm birth in the light of the newest data; an interested reader is also referred to other recent overviews on the same topic.

There have been significant strides made over the last two decades in the understanding of the natural history and pathophysiology of fetal thoracic lesions. Largely as a result of advances in prenatal ultrasound we are not only able to offer a tentative diagnosis of these lesions and advise parents about the prognosis but in selected centres offer the possibility of intervention for the most severely affected fetuses. Large gaps do remain in the current state of knowledge, in particular in the inability to predict pulmonary hypoplasia. Fetal surgery for lesions such as congenital cystic adenomatous malformation (CCAM) and pulmonary sequestration sequence (PSS) is still restricted to a very few centres and is reserved for those fetuses with hydrops where there is an extremely poor prognosis. Selection criteria for intervention have been refined over the last decade. Postnatal imaging studies, particularly computed tomography (CT) are essential in the evaluation of these lesions prior to surgical resection.

Despite many advances in the management of lung disorders of premature infants, respiratory distress syndrome (RDS) remains the major cause of morbidity and mortality in the preterm infant. Approximately 30% of infants born at less than 34 weeks gestation will develop RDS, secondary to deficiency in surfactant production by alveolar type II pneumocytes. Without surfactant, work of breathing is increased and gas exchange impaired by diffuse atelectasis, reduced pulmonary compliance and ventilation perfusion mismatch.

Kielland's forceps have been in obstetric practice for over 80 years but their use causes a wide spectrum of reactions in obstetricians. Those that have been well tutored in their use argue that they are a very effective instrument to achieve vaginal delivery in the malpositioned fetus, avoiding the problems of full cervical dilation caesarean section, with low complications in their hands. These exponents of the instrument argue that the “art” of obstetrics is demonstrated in the use of Kielland's forceps. However, others claim that Kielland's forceps are dangerous with high complication rates and that they should be confined to the obstetric museum. This paper reviews the history of the instrument and its inventor, to consider evidence for its effectiveness and its safety, to briefly consider other methods for delivery of the malpositioned fetus at full cervical dilation and finally to complete the journey by considering the future with particular emphasis on training the new generation of obstetricians.

Venous thromboembolism (VTE) is a leading cause of maternal mortality in the western developed world. VTE may present as deep vein thrombosis (DVT) or pulmonary embolism (PE) but if untreated can result in fatal PE. Although fatal PE is clearly the most significant consequence of VTE in pregnancy, DVT also often leads to morbidity related to the development of post-thrombotic syndrome (PTS). Pregnancy is an independent risk factor for VTE and the risk of VTE is 4–10 fold higher in pregnant women than in non-pregnant women of similar age. The puerperium represents a time of even higher risk. It is clear that many additional high-risk situations in pregnancy lower the threshold for thrombosis and warrant measures to prevent VTE and its complications. Risk factors for pregnancy-related VTE may be inherited or acquired. Acquired risk factors may be specifically related to the pregnancy or may have developed prior to pregnancy. Well-documented risk factors for pregnancy-related VTE include delivery by Caesarean section, previous VTE, and inherited or acquired thrombophilia. Other risk factors that have been identified include obesity, multiparity, multiple gestation, pre-eclampsia and medical conditions, such as sickle cell disease, that predispose to VTE. Our ability to diagnose VTE overall is poor as presenting signs and symptoms are extremely varied and unreliable. This is further complicated in pregnancy where signs and symptoms suggestive of VTE are common and invasive testing is more complicated. It is essential that physicians be vigilant in monitoring patients for the development of VTE and maintain a low threshold for considering thromboprophylaxis. Guidelines have recently been published by several medical societies to help with these difficult decisions. In this review the risk factors for the development of VTE in pregnancy will be discussed and guidelines for risk assessment presented. Management of patients who develop VTE in pregnancy is also outlined.

Hypertension is a common problem occuring in around 12–15% of all pregnancies. It remains one of the major causes of maternal death. Because of its perceived risk and the difficulty in deciding who has a problem and who does not, obstetricians often admit a patient found to be hypertensive to hospital for closer observation. This “play safe” approach can cause much inconvenience to the patient and her existing family. The result of this policy is that pregnancy hypertension is responsible for around 25% of all antenatal admissions. It is a major burden on health service resources in terms of manpower and hospital accommodation. Inpatient management may vary between different hospitals but the cornerstone is to keep the patient in hospital in order to allow serial monitoring of blood presure. Strict bedrest has been suggested to have therapeutic value but this has never been substantiated. As some of these patients will progress to more severe forms of the disease, it is important that there is early recognition of any signs of progression.

Major progress has been made in the knowledge and management of diabetes and pregnancy. However diabetes in pregnancy remains an important medical complication, with implications for mother and child in both the short and the long term. There are still controversies concerning the diagnostic criteria of gestational diabetes and concerning the best strategry for the treatment of this disorder. There is even less agreement about whether gestational diabetes is a pathological condition. The perinatal mortality in gestational diabetes has recently become as low as that in the general population. This has led to the conclusion that gestational diabetes is no more than a variant of metabolic adaptation during pregnancy. However, fetal hyperinsulinism is present in gestational diabetes, resulting in macrosomia, neonatal complications and most probably long term consequences. Improvements in the care of the preexisting insulin-dependent pregnant diabetic have been achieved but further progress in understanding the underlying mechanisms and in preventing and treating the disease are important goals for the near future.

Since the introduction of ultrasound into obstetrics during the 1960s, there has been rapid progress in the detection of genetic and nongenetic defects in utero. With the development of sampling procedures like amniocentesis, chorionic villus sampling (CVS) and fetal blood sampling, the obstetrician has been able to obtain fetal tissue and the parallel improvement in laboratory techniques has allowed the diagnosis of chromosomal anomalies and single gene defects from fetal cells. Amniocentesis and CVS have become well established techniques for routine prenatal diagnosis of chromosomal and metabolic disorders and fetal tissue is now accessible throughout all three trimesters.

Multiple pregnancy is associated with a high rate of perinatal loss – mainly due to preterm labour but with important contributions from fetal malformation, intrauterine growth retardation and twin-twin transfusion syndrome. The overall perinatal mortality rate is consistently around six times that of singleton pregnancies but the rate rises progressively with the number of fetuses. Rates of 63,164,200,214 and 416 per 1000 births have been recently reported for twins, triplets, quadruplets, quintuplets and sextuplets respectively. In addition to these alarming figures, it should be emphasized that the restricted concept of perinatal mortality obscures the real extent of loss. If we include late abortion (after 20 weeks), late neonatal deaths and deaths in infancy from perinatal causes, as well as the usual indices of perinatal mortality (stillbirths and early neonatal deaths) we find that the total loss rate from twin pregnancy alone doubles and may be close to 10%. Although the rate of loss from multiple pregnancies is now substantially higher than that associated with the pregnancies of diabetic women, the challenge of multiple pregnancy has not been met with the same commitment or organisation of specialized perinatal services as has diabetes.

The ductus venosus is a blood vessel functioning exclusively in the fetal circulation. According to many reports it was described for the first time by Giulio Cesare Arantius (Bologna 1530–1589) in his book De humano foetu libellus (1563) and it is often referred to as the ductus Arantii. Thorough research, however, has made it clear that Andreas Vesalius described this vessel in 1561 in his book Examen observationium Falloppii, which was published three years later. Moreover, it has been pointed out that the first and second editions of Arantius’s book did not contain a description of the ductus venosus but that this appeared in the 1571 edition.

We cannot talk to the fetus. Our ability to place the fetal condition accurately in context, therefore, lies with indirect evidence and inferred responses. We can examine the fetus. We can do so in a most comprehensive fashion, addressing the salient points of fetal anatomy, both structural and functional. We are able to perform routine evaluation of every system and the methodology is available to assess critical areas such as the heart and the central nervous system in specific detail.