Very early birth survival 'the same'

“Survival rates of babies born before the 24-week abortion limit have not improved significantly in the last decade,” The Daily Telegraph reports today. The newspaper says that a study found that survival rates of premature babies born at 23 weeks had not improved over the period that the study looked at, with just 18% leaving hospital. None of the babies born at 22 weeks survived.

The Guardian notes that this research has been published the week before the second reading of the human fertilisation and embryology Bill in the House of Commons. The House will discuss if the legal limit for abortion should be lowered from its current 24-week limit  to 20 weeks.

This well-conducted study appears to indicate that the survival rates of extremely premature babies born at 23 weeks did not improve between 1994 and 2005. This provides fresh evidence for the debate about whether or not the current legal time limits for aborting a healthy foetus need to be changed or not. The study is likely to stimulate further discussion on this most emotive of subjects.

Where did the story come from?

Professor of Neonatal Medicine David J. Field and colleagues from the University of Leicester and Nottingham City Hospital carried out the research. The study received support from NHS research and development funds supplied by healthcare commissioners in the Trent region. The study was published in the (peer-reviewed) British Medical Journal.

What kind of scientific study was this?

In this cohort study, the researchers investigated if there have been changes in survival rates for extremely premature infants who are born before the 26th week of pregnancy. They aimed to compare the survival rates from two five-year periods, 1994-1999 and 2000-2005, in a defined geographical region. The region examined was the Trent region in the UK, which has a population of about 4.6million people and about 55,000 births a year.

To do this, they used information from the Trent neonatal survey, which has registered all babies born prior to 32 weeks since 1990. All infants who were born before 26 weeks of pregnancy over the 10-year period in the Trent region were identified. To be included, infants had to have been alive at the onset of labour, therefore abortions were excluded.

The researchers looked at all outcomes of stillbirth or miscarriage, death before admission to neonatal intensive care, death while in intensive care and survival to home discharge. The neonatal survey covers the pregnancy, delivery and neonatal care details of infants. This includes the gestational age of the baby, which is established by the date of the mother’s last period, early or late dating scans or postnatal examination (considered to be the least reliable).

Information on stillbirths, miscarriages, and deaths prior to intensive care admission were obtained from the Confidential Enquiry into Stillbirths and Deaths in Infancy (CESDI), which contains information on all infant deaths after 22 weeks of pregnancy. The researchers used statistical methods to compare differences between the two periods.

What were the results of the study?

A total of 339,774 infants were born between 1994 and 1999 (855 prior to 26 weeks), and 317,473 born between 2000 and 2005 (797 prior to 26 weeks). There was no significant difference between the two periods in the number of babies born at 22 or 23 weeks who died in the delivery room (58% in 1994-1999 and 63% in 2000-2005). Deaths in the delivery room were lower for those infants delivered at 24 or 25 weeks, but there was still no significant difference between the two periods (13% in 1994-1999 and 10% in 2000-2005).

When the researchers looked at all infants delivered before 26 weeks, there was a significant improvement between the two periods in the number of babies surviving from intensive care to discharge (36% in 1994-1999 compared to 47% in 2000-2005). However, when they looked separately at babies born at 22 and 23 weeks and those born at 24 and 25 weeks it could be seen that the improvement was due to significantly improved survival rates in the later age groups
(24 weeks: from 24% in 1994-1999 to 41% in 2000-2005; 25 weeks: from 52% in 1994-1995 to 63% in 2000-2005), but not in the younger groups.

In both periods none of the infants born at 22 weeks survived to discharge, and there was no difference in numbers of those born at 23 weeks who survived to discharge in 1994-1999 (18.52%) compared to 2000-2005 (18.46%).

What interpretations did the researchers draw from these results?

The researchers concluded that, in the region of Trent, survival rates for infants born at 24 and 25 weeks have improved over the 10-year period. However, there has been no improvement in survival rates following intensive care admission for infants born at 23 weeks and continued unsuccessful care for all infants born at 22 weeks.

What does the NHS Knowledge Service make of this study?

This well-conducted study used reliable methods of data collection. It has given some quantitative evidence of the survival rates of extremely premature infants born before 26 weeks of pregnancy. There are a couple of points to note:

• The researchers acknowledge that they were unable to gain information on the situations surrounding the deaths of the infants; for example, deaths in the delivery room may have followed failed resuscitation attempts or a decision by the parents and professionals that resuscitation was inappropriate. Such details may have provided more insight as to whether medical advances in neonatal care have improved outcomes for extremely premature infants; for example, whether resuscitation attempts in an infant of 22 or 23 weeks are more likely to be successful now than they were a decade ago. However, they note that intensive care admissions and survival to discharge are unchanged, which suggests there has been little change in the delivery room setting.
• As the infant survival rates are from just one region they may not be the same elsewhere in the UK. However, the researchers say that the Trent region is “reasonably representative of the UK as a whole” and that their findings are unlikely to vary from those seen elsewhere. However, worldwide figures may not be the same, as Scandinavian studies have reported slightly higher survival rates for infants born at 23 weeks.
• The actual numbers of babies born at 22 and 23 weeks is fairly low (with a combined figure of 348 between 1994-1999 and 283 between 2000-2005). The smaller numbers may mean that significant differences between the two periods are more difficult to detect.

The research appears to indicate that the survival rates of extremely premature babies born at 23 weeks did not improve from 1994-2005. This provides fresh evidence to the debate about whether or not the current legal time limits for carrying out an abortion on a healthy foetus need to be changed or not. The study is likely to stimulate further discussion on this most emotive of subjects.

Sir Muir Gray adds…

This is important evidence in the debate about abortion, but the debate will be at least as much about values as about evidence.