Stillbirth and Lessons for Pregnancy Care

Stillbirth—death of a fetus in utero during the second half of pregnancy—has until recently been a neglected topic in perinatal medicine despite its frequent occurrence, conservatively estimated worldwide at more than 2.6 million deaths in 2008.¹ In the United States, fetal mortality is almost as common as infant mortality, affecting 1 in every 160 pregnancies.² Investigations into the causes of stillbirth and means to prevent it are receiving more attention.^{3 - 4} In 2006, the Eunice Kennedy Shriver National Institute of Child Health and Human Development funded the first multicenter prospective population-based study of adequate size of stillbirth in the United States with the hope of addressing important scientific gaps. This issue of JAMA includes 2 reports^{5 - 6} from the Stillbirth Collaborative Research Network (SCRN) that provide lessons about the causes of fetal death, opportunities to prevent stillbirth, and the implications of stillbirth research for all adverse pregnancy outcomes.

In a carefully constructed study that sought to capture all stillbirths in 5 defined geographic areas, the SCRN investigators enrolled every mother and stillborn neonate and also enrolled as controls 2 mothers and their liveborn neonate(s) (614 stillbirths and 1816 live births). Based on a comprehensive review of prenatal, intrapartum, and neonatal information, the authors confirmed known associations of stillbirth risk with maternal diabetes, obesity, smoking, multifetal pregnancy, and black race, and identified increased stillbirth risk for women pregnant for the first time or living alone. Risk factors evident at the beginning of pregnancy had low positive predictive value and were associated with relatively few stillbirths, limiting the opportunities for early intervention.

The level of detail enabled novel analyses that revealed 4 distinct categories of stillbirth based on gestational age, maternal race/ethnicity, and causes of death. Fetal deaths before 28 weeks' gestation included a group of fetal deaths between 20 and 24 weeks that occurred during labor and were more common among black women, and a second group in which fetal death from placental problems and other obstetrical conditions preceded labor. After 28 weeks' gestation, fetal deaths related to maternal hypertension predominated between 24 and 31 weeks, whereas cord abnormalities were the most common cause in the last weeks of pregnancy.

Fetal death during labor before 24 weeks' gestation is a consequence of extreme prematurity because interventions on behalf of the fetus are not indicated before viability. Black women, whose rate of preterm birth is double that of women from other racial/ethnic backgrounds, comprised a majority of this group. Stillbirth and preterm birth are often considered as separate entities, but this is strong evidence that the 2 are more closely linked than has been previously recognized. Additional evidence of common origins of stillbirth and preterm birth comes from the SCRN analyses of the risk of prior pregnancy outcomes for future stillbirth. As expected, a history of prior stillbirth was associated with increased risk (adjusted odds ratio, 5.9; 95% CI, 3.18-11.00). Increased risk of stillbirth (adjusted odds ratio, 3.13; 95% CI, 2.06-4.75) was also found in “nulliparous” women who had experienced a pregnancy loss before 20 weeks. These women are not parous in that they have not delivered a live or stillborn infant after 20 weeks' gestation, but their increased risk of subsequent stillbirth supports the contention that the traditional 20-week boundary between “miscarriage” and “birth” is not clinically or scientifically useful and thus should be abandoned.^{7 - 12}

Because pregnancy losses before and after 20 weeks' gestation have historically been thought to result from different pathways, research and vital statistics data have been collected and analyzed within that model. Recent reviews^{7 ,9} suggest otherwise, based on reports that stillbirths are associated with adverse outcomes in future pregnancies,¹³ as are second-trimester losses before and after 20 weeks.⁸ The SCRN reports add to the increasing realization that the current obstetrical taxonomy is an obstacle to improving pregnancy outcomes. Optimal investigation of the origins, treatment, and prevention of stillbirth and preterm birth will require data collected on deliveries before and after 20 weeks, regardless of whether the fetus is alive or not at the time of presentation for care.

Changes in the definitions and gestational age boundaries of preterm birth may seem simple to clinicians who do not work in perinatal medicine, but they are not. The “20-week line” and “liveborn vs stillborn” classifications are found in the vital statistics infrastructure and in the thinking of many obstetrics practitioners, researchers, and the general public. As the SCRN work highlights, pregnancies involving fetal death near the limit of viability, including those before 20 weeks, share important commonalities with pregnancies resulting in early preterm birth. As such, the reports from the SCRN will not only further the understanding of stillbirth but should also encourage the need to reframe thinking about how to address the problem of spontaneous preterm birth and the associated racial/ethnic disparities.