Premium
A functional definition of prolonged pregnancy based on daily fetal and neonatal mortality rates
Author(s) -
Divon M. Y.,
Ferber A.,
Sanderson M.,
Nisell H.,
Westgren M.
Publication year - 2004
Publication title -
ultrasound in obstetrics and gynecology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.202
H-Index - 141
eISSN - 1469-0705
pISSN - 0960-7692
DOI - 10.1002/uog.1053
Subject(s) - medicine , gestation , obstetrics , fetus , neonatal mortality , gestational age , infant mortality , pregnancy , odds ratio , odds , perinatal mortality , pediatrics , logistic regression , population , genetics , biology , environmental health
The standard definition of ‘prolonged pregnancy’ is 42 completed weeks of gestation or more. This definition is based on statistical considerations and is sanctioned by the American College of Obstetricians and Gynecologists (ACOG), the World Health Organization (WHO), and the International Federation of Gynecology and Obstetrics (FIGO)1–3. However, multiple studies have shown that the risk of adverse perinatal outcome increases as gestation advances beyond 41 weeks of gestation or earlier4–7. Thus it is clear that the definition of the upper limit of a normal pregnancy is somewhat arbitrary and certainly imprecise. Traditionally, perinatal mortality has been calculated per 1000 births. Using this methodology and a large computerized database, we evaluated weekly fetal and neonatal mortality rates in 181 524 accurately dated term and post-term pregnancies6. A statistically significant increase in fetal mortality was detected from 41 weeks’ gestation onward. In contrast, the odds ratios for neonatal mortality did not demonstrate a significant increase as gestational age (GA) advanced beyond 40 weeks. In view of the large number of women who are undelivered by 41 or even 42 weeks’ gestation, the data indicating that perinatal mortality is increased in these pregnancies present a major public health dilemma. Several publications have questioned the traditional method used to calculate fetal mortality7–12. As stated by Smith7, ‘estimating the probability of an event requires that the number of events (numerator) be divided by the number of subjects at risk for that event (denominator)’. Therefore, it seems logical to calculate fetal mortality as fetal deaths per 1000 on-going pregnancies (rather than per 1000 deliveries), whereas the probability of neonatal mortality should be calculated as the number of neonatal deaths per 1000 live births. Caughey et al.13 recently reached a similar conclusion. Indeed, as early as 1987, Yudkin et al.8 argued that the risk of stillbirth would be better measured as ‘the number of impending stillbirths divided by the total number of undelivered fetuses’. With this measure and a database of 40 635 deliveries (1978–1985), the authors showed an exponential increase in the risk of stillbirth, from 0.3/1000 at 29 weeks’ gestation to 1.8/1000 at 41 weeks’ gestation. Hilder et al.10 reported similar results in 1998. In both of these studies the authors calculated the weekly risk of fetal demise conditional on fetal survival until that gestational week8,10. One should not, however, ignore the fact that the fetus is exposed to the risk of intrauterine demise not only during the week of delivery, but also during the period leading to the week of delivery. The traditional conditional probability for stillbirth at a given GA fails to take into account the risks of death in all the preceding weeks. Therefore, Smith7 recently applied life-table analysis techniques and calculated the ‘weekto-week’ cumulative probabilities of stillbirth. This risk increased from approximately 0.4/1000 at 37 weeks’ gestation to 11.5/1000 at 43 weeks’ gestation. It is intuitively clear that the rate of fetal death and the rate of neonatal death can be viewed as two independent phenomena acting in opposite directions. For example, early in the third trimester of pregnancy, the risk of neonatal death (per 1000 live births) is very high, while