The terminology of pregnancy loss can be confusing and often varies among and even within countries. Traditionally, abortions (or miscarriages) refer to pregnancy losses prior to 20 weeks gestation, while fetal deaths or stillbirths are used to describe losses after 20 weeks gestation. The threshold of 20 weeks is somewhat arbitrary, and it may be more useful to classify pregnancy losses by stages of gestational development. For example, pregnancy losses could be classified as pre-embryonic or anembryonic (conception to 5 weeks gestation), embryonic (6 to 9 weeks gestation), and fetal (after 10 weeks gestation) losses. In the past, anembryonic losses were termed blighted ova. This term is best avoided. Another approach would be to describe losses prior to 20 weeks gestation as early losses and those after 20 weeks gestation as late losses.

It is important to specify the timing in gestation and nature of pregnancy loss as accurately as possible. This can be difficult because death or failure of growth of the conceptus may precede clinical symptoms of miscarriage by days or weeks. Thus, ultrasound findings, histologic examination, and human chorionic gonadotropin (hCG) levels often are more useful than clinical data when characterizing pregnancy losses. The etiologies of pregnancy losses vary over gestation. Genetic problems are more common in pre-embryonic or anembryonic losses, while antiphospholipid syndrome (APS) and heritable thrombophilias are more likely in losses after 10 weeks gestation. Also, the recurrence risk of pregnancy loss is influenced by gestational age. Pregnancy losses tend to recur at similar times during gestation, and in patients with recurrent pregnancy loss, those with fetal deaths after 10 weeks gestation have worse subsequent pregnancy outcomes than women with recurrent early losses.

Stillbirth often is defined as the death of a fetus after 20 completed weeks of gestation. In cases of uncertain gestational age, losses weighing >500 g are considered
stillbirths. However, others believe that a more useful definition would include death of a fetus past the age of viability. This has prompted the use of 24 or even 28 weeks gestation as a threshold for defining stillbirth. Another controversial issue is whether or not to include both antepartum and intrapartum deaths as stillbirths. Some but not all studies distinguish between the two. This is an important distinction, as causes and strategies to reduce antepartum and intrapartum losses are quite different. Intentional fetal death (e.g., pregnancy termination in the setting of fetal anomalies or preterm premature rupture of membranes) poses another problem for vital statistics. Pregnancy terminations often are excluded when counting stillbirths, but the issue remains controversial. It seems appropriate to exclude terminations as a distinct entity. However, in many cases, these fetuses would be stillborn if the pregnancy were allowed to continue. As such, their exclusion might alter the true impact of stillbirth on a population.

In 2003, the stillbirth rate (after 20 weeks gestation) in the United States was 6.23 per 1,000 live births. This represents a slight and steady decline since 1990, when the rate was over 7.5 per 1,000 live births (Fig. 23.1). Just over one half of stillbirths in the United States occur between 20 and 27 weeks gestation. It is important to note that this may underestimate the true rate, as stillbirths are likely underreported. The scope of the problem is considerably higher in many parts of the world. Stillbirth rates are estimated to range from 5 per 1,000 births in rich countries to 32 per 1,000 births in southern Asia and sub-Saharan Africa. The estimated number of global stillbirths per year is 3.2 million.

In the United States, stillbirth rates continue to be increased in non-Hispanic black women. In 2003, the fetal mortality rate for non-Hispanic blacks was 11.56 per 1,000 births compared with 4.94 per 1,000 births for non-Hispanic whites. Fetal mortality also is relatively higher for teenagers, unmarried women, and women over 35 years of age.

It is difficult to obtain an accurate estimate of the fetal death rate between 10 and 20 weeks gestation, because such data are not routinely collected. According to the National Survey of Family Growth, estimates of total fetal losses per year in the United States are about one million. The majority of these occur prior to 20 weeks gestation.

Numerous classification schemes have been used to catalog stillbirths based on etiology. This has proved difficult for several reasons. First, it often is difficult to be certain of a “cause” of death. Many risk factors such as diabetes, obesity, or heritable thrombophilias are associated with an increased risk of fetal death. However, the vast majority of women with these risk factors have live-born infants. Thus, it is difficult to be certain that these factors led to the stillbirth. Second, there may be more than one potential cause of stillbirth in the same patient. For example, what would be considered the etiology in a fetus with trisomy 18 that also has evidence of bacterial infection? Stillbirth may be due to the additive or interactive effects of several disorders. Finally, a cause of death often is never determined. In many cases, this is due to a lack of systematic evaluation into potential etiologies. However, a cause of death may not be found despite a comprehensive evaluation. This is especially true in cases of term stillbirth.

Because of these problems, no single classification scheme has been universally adopted, and new ones are being developed. Additional confusion arises from the use of different definitions for fetal death, the occasional inclusion of neonatal death, and different standards regarding causes versus associations in the classification schemes. The Wigglesworth classification probably is in most widespread use today (Table 23.1). A problem with all of the classification systems is that many stillbirths remain unexplained. Recently, Gardosi and colleagues introduced a new system that explains a much larger proportion of fetal deaths than prior schemes (Table 23.2). However, many of the etiologies in the new system may be risk factors such as small-for-gestational-age fetus (SGA) rather than etiologies. Forthcoming international conferences hopefully will facilitate the development of uniform definitions and classification systems for fetal death.