INTRODUCTION
Stillbirth is a devastating event and yet it harbours an up to 22-fold
recurrence risk in future
pregnancies.1,
2 With the aim to reduce the annual
stillbirth rate worldwide, the ability to predict the likelihood of such
event by accurate risk stratification may prompt both parents and
clinicians to embark upon a suitable and targeted antenatal surveillance
program.
Prediction models for stillbirths are most commonly defined as “models,
scores or clinical decision tools” which aid in estimating the risk of
stillbirth in a pregnant woman based upon certain
variables.3 In a recent
review, the most commonly used variables in prediction models for
stillbirths have been identified to be maternal age, body mass index
(BMI) and maternal diabetes, yet strongest evidence of association with
stillbirth was for nulliparity, pre-existing hypertension and maternal
obesity.4 As about 11.2
to 64.9% of stillbirths in high income countries are due to placental
dysfunction,5 a triad of
the latter factors is most likely contributing to such. Whilst the
pathomechanisms of the individual risk denominators might work
differently on the axis leading to fetal death, the synthesis of these
variables into a prediction model is helpful for early recognition and
intervention to prevent adverse perinatal outcome. To date, 69
prediction models for stillbirths have been described in
literature.3
By this study we aim to apply the demographic model of the Fetal
Medicine Foundation (FMF) Stillbirth Risk
Calculator6 based upon
maternal characteristics (weight, ethnicity and smoking), medical
history [diabetes, chronic hypertension, systemic lupus erythematosus
(SLE) and anti-phospholipid syndrome (APS)] and obstetric history
(parity, stillbirth and/or preeclampsia in previous
pregnancies)7 in our
single-centre cohort of intrauterine fetal deaths (IUFD) and matched
live births as an independent dataset for external validation of this
prediction tool.