Stillbirths and infant deaths

Research led by the University of Glasgow has found that widening the definition of ‘at risk’ babies based on predicted birth weight may reduce the number of stillbirths and infant deaths.

The study, which is published in PLOS Medicine, found that birth weight at or below the 25th centile, or at or above the 85th centile is associated with greater risk of adverse outcomes such as stillbirths and infant deaths.

Closer surveillance, or earlier delivery of babies with these predicted birth weight may reduce these adverse outcomes.

In developed countries, like the UK, one-third of stillbirths and infant deaths still occur at full term babies. No consensus currently exists about what defines a small or large foetus or infant at term and what birthweight thresholds should trigger surveillance and/or intervention.

The research, led by Dr Stamatina lliodromiti from the School of Medicine, used data from 979,912 term singleton pregnancies (a child born singly, rather than part of a multiple birth) over a 19-year period in Scotland, and externally validated the findings in an independent UK cohort including 10,515 pregnancies.

The researchers, who included colleagues from the University of Bristol and the University of Cambridge, studied the associations of birth weight centiles with stillbirth, infant mortality, admission to the neonatal unit and low Apgar score (an assessment of a baby’s overall well-being at birth).

Dr Iliodromiti said: “Currently babies that are predicted to be too small (most frequently less than the 10th centile) or too large (most frequently over the 90th centile) are considered at risk.

“We provide convincing evidence that how we define whether a baby is too small or too large needs to be re-visited. We anticipate that by redefining this, we can continue to reduce stillbirths.”

In fact, customised birth weight centiles (accounting for sex, gestation, and maternal characteristics) are increasingly being adopted by many maternity units. However it has not yet been tested whether these predict adverse outcomes better.

Although they were not able to assess fully customised centiles, the researchers did look at whether partially customised centiles (those only allowing for maternal height and parity) perform better in predicting adverse outcomes. However they found that these partially customised centiles do not improve prediction performance.

Dr lliodromiti added: “Replication of the analysis with fully customised birth weight centiles that additionally account for maternal ethnicity and weight is required. We can then definitively answer whether customised birthweight centiles are beneficial.”

The paper, Customised and Noncustomised Birth Weight Centiles and Prediction of Stillbirth and Infant Mortality and Morbidity: A Cohort Study of 979,912 Term Singleton Pregnancies in Scotland is published in PLOS Medicine. The study is funded by the MRC.

Recent Issues