Behind The Medical Headlines

When is the ideal time for a pre-term fetus to be delivered?

• Dr IA Laing, Consultant Neonatologist, Simpson Centre for Reproductive Health, Royal Infirmary of Edinburgh, Edinburgh, Scotland

Summary

It can be very difficult to decide if it would be better to deliver a compromised, or at-risk, fetus early, and at a developmental stage when there may be a higher risk of death or disability, or to delay until there is no clinical doubt that delivery is necessary and the risks may be lower. In this article Dr Ian Laing provides an overview of some recent studies which have tried to answer this difficult question.

Declaration of interests: No conflict of interests declared

In August 2004, The Lancet published data on brain development in survivors of the multicentred Growth Restriction Intervention Trial (GRIT). The aim of this study¹ was to identify compromised fetuses between 24 and 36 weeks gestation and answer the question whether it was safer to deliver them immediately or to delay until there was no clinical doubt that delivery was necessary. Five-hundred and eighty-eight such fetuses were identified in 69 hospitals in 13 European countries. Two-hundred and ninety-six were randomly assigned to immediate delivery and 292 to delayed delivery. The median interval between randomisation and delivery in the immediate delivery group was 0·9days, and 4·9 days in the delay group. The authors had previously reported on the stillbirths and neonatal deaths in these pregnancies², and identified similar mortality before discharge between immediate and delayed delivery groups.

The main outcome measures of the current report were death and disability at or beyond two years of age. These outcome measures were recorded in 98% of immediate births and 97% of delayed births. Overall rates of death and disability were similar: 19% of immediate births and 16% of delayed births. In the subgroup of infants born at gestation less than 31 weeks, the outcomes appeared to improve by delaying the delivery where possible.

Obstetricians and midwives have the unenviable task of looking after patients whom they cannot see. The mother’s health is often considered to be their primary responsibility, but both professions have the added burden of caring for single or multiple fetuses, and can monitor their progress only indirectly. The wellbeing of the pregnancy can be judged by the mother’s health, her own weight gain and the absence of acute maternal disease (including hypertension or vaginal bleeding). In addition, fetal size can be judged by the obstetrician’s hands, and the heart rate monitored by auscultation. Biochemical tests such as urinary oestriol concentrations have failed to live up to the promise of monitoring placental function and have largely been abandoned. Ultrasound and heart rate monitoring by cardiotocography help assess fetal wellbeing more effectively.

The challenge of monitoring fetal health is particularly serious in the presence of growth-restriction. Such a fetus has few metabolic reserves, and sudden death in pregnancy may be unheralded. Labour is an intermittently hypoxic event, and anaerobic metabolism may not be an option where stores of fat and glycogen are inadequate. More recently placental and fetal arterial Doppler flow-velocity waveforms have guided decision making. These have proved particularly effective in assessing the growth-restricted pregnancy and are a useful adjunct in assessment of the very preterm fetus where cardiotocographical monitoring may be unhelpful.

The GRIT study is in many ways an astonishing achievement, representing as it does the gathering of data in 13 countries over a seven-year period. It implies that in 548 situations the obstetrician found himself or herself unable to decide whether to deliver immediately or not. The success of follow-up indicates excellent organisation, and the paediatricians describing outcomes did not know to which study group an individual fetus had been allocated.

From a neonatologist’s point of view, the GRIT study group chose an unexpected range of gestational ages to study. The 24-week gestation baby is very different from that at 36 weeks. In the absence of severe congenital abnormalities, the current infant mortality after 32 weeks’ gestation is low: the causes of this rare event include asphyxia, necrotising enterocolitis and infection. Because of the availability of intratracheal replacement surfactant, significant respiratory distress syndrome is rare in this group. Obstetricians should therefore have a low threshold for delivery in this relatively mature group if the fetus appears to show signs of compromise. By contrast, before 32 weeks, and particularly in the extreme preterm infant, there is a much higher mortality, and the levels of morbidity have recently been emphasised in the Epicure Study,³ in which 49% of surviving infants born at less than 26 weeks’ gestation had some disability at 30 months of age and 19% were severely disabled.

It is in this immature group of infants that one might expect to find benefits in delaying time of delivery. With every week that passes there should be a decreasing occurrence of serious complications including intraventricular haemorrhage, retinopathy of prematurity and sepsis. On the other hand, delay may expose the growth-restricted fetus to ischaemic injury of the brain, resulting in asphyxia, periventricular leucomalacia and intraventricular haemorrhage, as well as the significant risk of intra-uterine death. This surely is the group where the main challenge lies. The authors of the GRIT study did examine a subgroup of infants at 24–30 weeks’ gestation and demonstrated that the immediate delivery group had higher cerebral palsy rates (10% vs 0%) and more infants with a Griffiths Developmental Quotient of less than 71 (10% vs 4%), but the numbers were too small to achieve statistical significance. Regrettably sample size is often the limiting factor in such epidemiological studies.

In essence this study has shown a small increase in fetal death if the obstetrician delays delivery, and a small increase in neonatal death if early delivery is chosen. The most important information from this study may emerge from detailed follow-up of the surviving children.

Nevertheless this was an admirable trial, even though it does not inform obstetricians about when, in an individual case, the optimal time of delivery should be. The art of clinical medicine is once again found to be paramount, until a new and reliable method of fetal monitoring can be found.

ADDENDUM – WHEN IS THE IDEAL TIME FOR A PRETERM FETUS TO BE DELIVERED?

Addendum

An important new British study linked to this issue has been published in the New England Journal of Medicine (NEJM).⁴ This study provides data on the outcome of survivors of the EPICure study⁵ at school age. These new data provide a new perspective on the future prospects of infants born at a gestation of less than 26 weeks.

Findings

The EPICure study describes 1,185 infants born between 22 and 25 weeks’ gestation. Three-hundred and fourteen survived to discharge home, and a further 6 died thereafter. Sixty-seven were then lost to follow-up, and so 241 were assessed at a median age of six years and four months, using 160 classmates delivered at full term as a comparison group. Forty-one per cent of survivors had cognitive deficits. The rates of severe, moderate and mild disability were 22%, 24% and 34%, respectively, although disabling cerebral palsy was present in only 30 children (12%). Of the 63 children with severe disability at 30 months of age, 54 (86%) still had moderate-to-severe disability at school age. In contrast, other disabilities identified at the age of 30 months were poorly predictive of developmental problems at six years of age.

Comment

There are some important conclusions to draw from these data. Forty-four percent of infants born at 25 weeks’ gestation survived to discharge, whereas delivery at 22 weeks almost universally resulted in neonatal death. Many units in the United Kingdom do not offer resuscitation to infants born at 22 weeks, and many of us as clinicians are reluctant to provide invasive intensive care at 23 weeks’ gestation unless fully informed parents exhort us to do so.

Neonatologists, obstetricians, midwives and parents must increasingly recognise that infants born less than 25 weeks’ gestation who survive are at risk of disability at school age. In the EPICure study only 20% were totally free of disability at school age and so a guarded prognosis must be provided. On the other hand we must be clear about definitions in our interpretation of the data.

1. ‘Severe disability’ was employed if it was considered the child was likely to be highly dependent on care-givers, e.g. non-ambulant cerebral palsy, profound hearing loss or blindness
2. ‘Moderate disability’ referred to children who were likely to be reasonably independent, e.g. ambulant cerebral palsy, some hearing loss, some visual impairment
3. ‘Mild disability’ included neurological signs with minimal functional consequences

It can be stated that over half the survivors had mild disability or no disability at school age. In addition, some of the 24% with moderate disability were improved with spectacles and hearing aids. For individual families such outcomes may be regarded as very positive. All of us can take some encouragement from these figures.

Finally, it is essential to recognise that these crucial follow-up data refer to a large cohort of infants born in 1995, and by definition they tell us much about the standard of perinatal care 10 years ago. There have been many changes in Perinatology since then, some of which may be expected to improve outcomes still further. These include the widespread use of prophylactic surfactant against respiratory distress syndrome, the virtual abandonment of postnatal steroids which have been shown to be associated with the development of cerebral palsy, and increasingly sophisticated trend monitoring.

Clinicians should be cautious in managing infants born at the threshold of viability. Nevertheless, the majority of survivors may be considered to have a fulfilling life, and we must maintain our enthusiasm for providing this vulnerable group with the highest quality care.

References

1. The GRIT study group. Infant wellbeing at 2 years of age in the Growth Restriction Intervention Trial (GRIT): multicentred randomized controlled trial. Lancet 2004; 364:513–20.
2. GRIT Study Group. A randomized trial of timed delivery for the compromised preterm fetus: short term outcomes and Bayesian interpretation. BJOG 2003; 110:27–32.
3. EPICure Study Group. Neurologic and developmental disability after extremely preterm birth. N Engl J Med 2000; 343(6):429–30.
4. Marlow N, Wolke D, Bracewell MA, Samara M for the EPICure Study Group. Neurologic and developmental disability at six years of age after extremely preterm birth. N Engl J Med 2005; 352:9–19.
5. EPICure Study Group. Neurologic and developmental disability after extremely preterm birth. N Engl J Med 2000; 343(6):429–30.