Prediction of perinatal mortality at an early stage of pregnancy

AIM: To define the probability of foetal and neonatal death (PFND) in Lithuania by means of the Medical Birth registry. METHODS: The Medical Birth registry for the year 2002 (29,619 pregnancies resulting in delivery) was used, covering social, environmental, health behaviour factors, obstetric history, health status of pregnant women, and delivery. RESULTS: A statistically significant model has been obtained in which 12 out of 45 analysed factors were significant for the risk of perinatal mortality (PM). In the case of previous neonatal pathology, e.g. damage to CNS, the risk of PM (OR 5.2; 95% CI = 2.1-13.3) was similar to the reported manual work or harmful working conditions during pregnancy (OR 4.8; 95% CI = 1.9-12.1 and OR 4.5; 95% CI = 1.3-15.2, respectively). The influence of father's alcohol abuse (OR 2.6; 95% CI = 1.4-4.7) was comparable to that of the stillborns or premature deliveries. The combination of several factors increased PFND significantly: maternal smoking during pregnancy and paternal smoking and alcohol abuse raised PFND to 0.03 (OR 6.8), mother's hard manual work in combination with harmful conditions during pregnancy increased OR more than 10 times (PFND 0.32) as compared with the situation when only one of those factors was present. CONCLUSIONS: The influence of social, environmental, and behavioural factors on the risk of PM was comparable to that of some of the complications during a previous pregnancy. The combination of risk factors significantly increased PFND. Using modelling, the individual prognosis may be presented to every pregnant woman at an early stage of pregnancy to potentially decrease the impact of unfavourable risk factors on PM.