Research adds insight on stillbirth risk factors

Stillbirth is a major public health problem, occurring in approximately 1 of every 160 pregnancies in the United States. The rate has remained stagnant since 2006. Prior to that time, from 1990 to 2006, the rate declined somewhat, but only half as much as the decline in infant mortality during this time period. Racial disparities also have persisted, with non-Hispanic black women having more than a twofold increase in risk (Natl Vital Stat Rep. 2012;60:1-22).

Research conducted by the Stillbirth Collaborative Research Network (SCRN) and others has provided us with insight on risk factors and on probable and possible causes of death among stillbirths, which are defined as fetal deaths at 20 or more weeks’ gestation. We know from SCRN data, for instance, that black women are more likely to have stillbirths associated with obstetric complications and infections than white and Hispanic women. However, we still cannot explain a substantial proportion of stillbirths, despite a complete evaluation, or predict who will have a stillbirth.

What we can do as obstetricians is be aware that stillbirth is one of the most common adverse pregnancy outcomes in the United States and counsel women regarding risk factors that are modifiable. Moreover, when stillbirth happens, a complete postmortem evaluation that includes autopsy, placental pathology, karyotype or microarray analysis, and fetal-maternal hemorrhage testing is recommended (Obstet Gynecol. 2009;113[3]:748-61). Recent data show that each of these four components is valuable and should be considered the basic work-up for stillbirth.

Risks and causes

Pregnancy history was the strongest baseline risk factor for stillbirth in an analysis of 614 stillbirths and 1,816 live births in the SCRN’s population-based, case-control study conducted between 2006 and 2008. The SCRN was initiated by the Eunice Kennedy Shriver National Institute of Child Health and Human Development in 2003. This critical population-based study was conducted at 59 U.S. tertiary care and community hospitals in five catchment areas and has been analyzed in more than 15 published reports.

Women with a previous stillbirth have been known to be at 5- to 10-fold increased risk of a recurrence of stillbirth, and the SCRN findings confirmed this. The study added to our knowledge, however, with the finding that even a prior pregnancy loss at less than 20 weeks’ gestation increased the risk for stillbirth.

Other risk factors identified in the study, in addition to race, included having a multifetal pregnancy (adjusted odds ratio of 4.59), diabetes (AOR of 2.50), maternal age of 40 years or older (AOR of 2.41), maternal AB blood type (AOR of 1.96, compared with type O), a history of drug addiction (AOR of 2.08), smoking during the 3 months prior to pregnancy (AOR of 1.55-1.57, depending on amount), and being unmarried and not cohabitating (AOR of 1.69). Regarding racial disparity, the study showed that elevated risk of stillbirth for non-Hispanic blacks occurred predominantly prior to 24 weeks of gestation.

As in prior research, overweight and obesity also conferred elevated risks in the SCRN study (AORs of 1.43 and 1.72, respectively), and these risks were not explained by either diabetes or hypertension (JAMA. 2011;306:2469-79).

The use of assisted reproductive technology was not included in the study’s multivariate model, but previous research has shown a fourfold increased risk of stillbirth for singleton IVF/ICSI pregnancies. The reason is unclear, but the risk appears to be more related to IVF/ICSI rather than the underlying infertility (Hum Reprod. 2010 May;25[5]:1312-6).

A previous preterm or small-for-gestational-age birth has also been shown in prior research to be a significant risk factor for stillbirth. Less clear is the role of previous cesarean delivery in stillbirth risk. An association has been demonstrated in several studies, however, including one involving about 180,000 singleton pregnancies of 23 or more weeks’ gestation. Women in this cohort who had a previous cesarean delivery had a 1.3-fold increased risk of antepartum stillbirth, after controlling for important factors such as race, body mass index (BMI), and maternal disease (Obstet Gynecol. 2010 Nov;116[5]:1119-26).

In another analysis of the SCRN study looking specifically at causes of stillbirth, a “probable” cause of death was found in 61% of cases and a “possible or probable” cause of death in more than 76% of cases. The most common causes were obstetric complications (29.3%), placental abnormalities (23.6%), fetal genetic/structural abnormalities (13.7%), infection (12.9%), umbilical cord abnormalities (10.4%), hypertensive disorders (9.2%), and other maternal medical conditions (7.8%).

A higher proportion of stillbirths in non-Hispanic black women, compared with non-Hispanic white women and Hispanic women was associated with obstetric complications (43.5%) and infections (25.2%). This finding combined with the finding that stillbirth in black women often occurs at less than 24 weeks’ gestation suggests that measures aimed at reducing the rate of spontaneous preterm birth in black women could potentially reduce the rate of stillbirth as well (JAMA. 2011 Dec 14;306[22]:2459-68).

Work-up and prevention

Prevention of stillbirth requires that we identify the women at highest risk, and thus far this ability still eludes us. Apart from occurrence of previous stillbirth or pregnancy loss, other risk factors have had limited predictive value in the SCRN analyses and other research.

Biomarkers such as a low PAPP-A during the first trimester and a high AFP in the second trimester – as well as Doppler imaging of the uterine artery – have also been associated with stillbirth, but again, the positive predictive value has been shown to be low (Clin Obstet Gynecol. 2010 Sep;53[3]:597-606). More research is needed to determine if some combination of biochemical markers, imaging, and other risk factors can predict which women are at highest risk.

In the meantime, attention can be paid – in the preconception period if possible – to modifiable risk factors such as maternal obesity, diabetes, and smoking. About 10% of stillbirths are associated with maternal conditions such as hypertension and diabetes, and late stillbirths in particular (28 weeks or later) are associated with maternal medical conditions that are potentially preventable.

Normalization of prepregnancy weight should be a goal, since the overall risk of stillbirth appears to increase independently with increasing BMI. Glycemic control should also be achieved: A recent meta-analysis of preconception and prenatal care of diabetic women estimated “conservatively” that 10% of diabetes-associated stillbirths could be prevented with early detection and glycemic control (BMC Public Health. 2011;11 Suppl 3:S2). Research has also shown that women who quit smoking between their first and second pregnancy reduce their stillbirth risk to that of nonsmokers in the second pregnancy (BJOG. 2007 Jun;114[6]:699-704).

When stillbirth happens, a thorough work-up is recommended in order to counsel for future pregnancies and decrease the risk of recurrence. Evaluations for causes of stillbirth are too often incomplete in the United States for various reasons, including emotional, cultural, and resource factors. Even if a cause is not found, many families appreciate knowing that every effort has been made to determine a cause of death.

Four components of evaluation – autopsy, placental examination, karyotype or microarray analysis, and fetal-maternal hemorrhage testing – have proven to be high-yield tests when performed in all cases of stillbirth.

In the SCRN study, of 512 stillbirths undergoing a complete evaluation, 66.4% had a positive result – defined as abnormalities contributing to a probable or possible cause – for at least one of the first three tests (JAMA. 2011 Dec 14;306[22]:2459-68).

A Dutch study of 1,025 stillbirths similarly demonstrated that all four tests are justified. A test was defined as valuable in this study if it established or excluded a cause of stillbirth. Placental examination was determined to be the most valuable test, helping to determine a cause of death in 95.7% of cases. Autopsy was valuable 72.6% of the time, and cytogenetic analysis was valuable in 29% of cases.

Kleihauer-Betke testing for fetal-maternal hemorrhage was positive in 11.9% of women. However, fetal maternal hemorrhage was considered the cause of death in only 1.3%.of cases because, beyond a positive Kleihauer-Betke test, evidence of fetal anemia confirmed by placental examination and/or autopsy was required for hemorrhage to be considered the cause of death (Am. J. Obstet. Gynecol. 2012;206:53.e1-12). Because Kleihauer-Betke testing is ideally performed before induction, authors of both the SCRN study and the Dutch study believe it is a valuable test to be offered in all cases.

In both studies, the yield of other stillbirth diagnostic tests (for example, maternal serology, hormone assessment, and toxicology screen) was low, indicating that these tests are considered sequential and can be performed only when the clinical history or findings of the four core tests raise suspicion of particular potential causes. Antinuclear antibody testing and TORCH (toxoplasmosis, rubella, cytomegalovirus, herpes simplex) titers have an extremely low yield and are generally not useful.

For detecting genetic abnormalities after stillbirth, it appears that microarray analysis is superior to karyotype analysis. In a SCRN analysis of samples from 532 stillbirths, microarray yielded results more often and identified more genetic abnormalities. Unlike karyotype, it does not require live cells, which makes it preferable for stillbirth evaluation (N Engl J Med. 2012 Dec 6;367[23]:2185-93).

Current research

One of the more significant studies underway on prevention is looking at labor induction as an intervention for reducing stillbirths and improving other perinatal outcomes. The ARRIVE trial (“A Randomized Trial of Induction Versus Expectant Management”), currently in the recruitment stage, will examine outcomes after induction at 39 weeks’ gestation, compared with expectant management in 6,000 patients (clinicaltrials.gov/ct2/show/NCT01990612).

Common wisdom informed by retrospective cohort studies has long told us that inducing labor prior to 41 weeks’ gestation is associated with a higher risk of cesarean delivery in nulliparous women. However, recent observational data have suggested that women whose labor is induced actually have fewer cesarean deliveries and better perinatal outcomes, including a lower risk of stillbirth (AJOG 2012;207:502.e1-8).

In addition, a meta-analysis published in 2014, as the ARRIVE trial was taking shape, reported a 12% reduction in cesarean delivery, and a reduced risk of stillbirth, among women whose labor was induced. The initial cervical score did not impact the main findings (CMAJ. 2014 Jun 10;186[9]:665-73). If these findings are confirmed in the ARRIVE trial, we could see a new opportunity for stillbirth prevention.

Another ongoing study of 10,000 singleton pregnancies – the Nulliparous Pregnancy Outcomes: Monitoring Mothers-to-Be (nuMoM2b) study – may also lead to prevention strategies in women for whom the current pregnancy will lead to their first delivery. Among the questions being examined in this eight-site study are whether sleep-disordered breathing, or apnea, and a supine sleep position are risk factors for adverse pregnancy outcomes including stillbirth.

Supine sleeping in the last month of pregnancy was strongly associated with stillbirth in a recent analysis from the Sydney Stillbirth Study (Obstet Gynecol. 2015 Feb;125[2]:347-55), and an early analysis of a nuMoM2b subset has shown associations between sleep-disordered breathing in midpregnancy and the development of hypertensive disorders of pregnancy, and between sleep-disordered breathing in early- and mid-pregnancy and gestational diabetes (Am J Obstet Gynecol. 2015;212:S424-425).

The possible role of low-dose aspirin in preventing stillbirth also needs more exploration. A recent randomized trial of women attempting to become pregnant after having had one or two prior pregnancy losses found no difference overall in live birth rates between those who took low-dose aspirin and those assigned to placebo. However, there was one subgroup – women with a single loss at less than 20 weeks’ gestation during the previous year – in which live birth rates were higher in the aspirin group (Lancet. 2014 Jul 5;384[9937]:29-36). More research is necessary to determine if low-dose aspirin administration in women with a previous stillbirth improves pregnancy outcome.

Dr. Reddy is a member at the Pregnancy and Perinatology Branch of the Eunice Kennedy Shriver National Institute of Child Health and Human Development. She is a board-certified ob.gyn. and maternal-fetal medicine specialist. She is the program scientist for the Maternal-Fetal Medicine Units Network and for the Stillbirth Collaborative Research Network. The comments and views of the author do not necessarily represent the views of the NICHD.

Stillbirth is a major public health problem, occurring in approximately 1 of every 160 pregnancies in the United States. The rate has remained stagnant since 2006. Prior to that time, from 1990 to 2006, the rate declined somewhat, but only half as much as the decline in infant mortality during this time period. Racial disparities also have persisted, with non-Hispanic black women having more than a twofold increase in risk (Natl Vital Stat Rep. 2012;60:1-22).

Research conducted by the Stillbirth Collaborative Research Network (SCRN) and others has provided us with insight on risk factors and on probable and possible causes of death among stillbirths, which are defined as fetal deaths at 20 or more weeks’ gestation. We know from SCRN data, for instance, that black women are more likely to have stillbirths associated with obstetric complications and infections than white and Hispanic women. However, we still cannot explain a substantial proportion of stillbirths, despite a complete evaluation, or predict who will have a stillbirth.

What we can do as obstetricians is be aware that stillbirth is one of the most common adverse pregnancy outcomes in the United States and counsel women regarding risk factors that are modifiable. Moreover, when stillbirth happens, a complete postmortem evaluation that includes autopsy, placental pathology, karyotype or microarray analysis, and fetal-maternal hemorrhage testing is recommended (Obstet Gynecol. 2009;113[3]:748-61). Recent data show that each of these four components is valuable and should be considered the basic work-up for stillbirth.

Risks and causes

Pregnancy history was the strongest baseline risk factor for stillbirth in an analysis of 614 stillbirths and 1,816 live births in the SCRN’s population-based, case-control study conducted between 2006 and 2008. The SCRN was initiated by the Eunice Kennedy Shriver National Institute of Child Health and Human Development in 2003. This critical population-based study was conducted at 59 U.S. tertiary care and community hospitals in five catchment areas and has been analyzed in more than 15 published reports.

Women with a previous stillbirth have been known to be at 5- to 10-fold increased risk of a recurrence of stillbirth, and the SCRN findings confirmed this. The study added to our knowledge, however, with the finding that even a prior pregnancy loss at less than 20 weeks’ gestation increased the risk for stillbirth.

Other risk factors identified in the study, in addition to race, included having a multifetal pregnancy (adjusted odds ratio of 4.59), diabetes (AOR of 2.50), maternal age of 40 years or older (AOR of 2.41), maternal AB blood type (AOR of 1.96, compared with type O), a history of drug addiction (AOR of 2.08), smoking during the 3 months prior to pregnancy (AOR of 1.55-1.57, depending on amount), and being unmarried and not cohabitating (AOR of 1.69). Regarding racial disparity, the study showed that elevated risk of stillbirth for non-Hispanic blacks occurred predominantly prior to 24 weeks of gestation.

As in prior research, overweight and obesity also conferred elevated risks in the SCRN study (AORs of 1.43 and 1.72, respectively), and these risks were not explained by either diabetes or hypertension (JAMA. 2011;306:2469-79).

The use of assisted reproductive technology was not included in the study’s multivariate model, but previous research has shown a fourfold increased risk of stillbirth for singleton IVF/ICSI pregnancies. The reason is unclear, but the risk appears to be more related to IVF/ICSI rather than the underlying infertility (Hum Reprod. 2010 May;25[5]:1312-6).

A previous preterm or small-for-gestational-age birth has also been shown in prior research to be a significant risk factor for stillbirth. Less clear is the role of previous cesarean delivery in stillbirth risk. An association has been demonstrated in several studies, however, including one involving about 180,000 singleton pregnancies of 23 or more weeks’ gestation. Women in this cohort who had a previous cesarean delivery had a 1.3-fold increased risk of antepartum stillbirth, after controlling for important factors such as race, body mass index (BMI), and maternal disease (Obstet Gynecol. 2010 Nov;116[5]:1119-26).

In another analysis of the SCRN study looking specifically at causes of stillbirth, a “probable” cause of death was found in 61% of cases and a “possible or probable” cause of death in more than 76% of cases. The most common causes were obstetric complications (29.3%), placental abnormalities (23.6%), fetal genetic/structural abnormalities (13.7%), infection (12.9%), umbilical cord abnormalities (10.4%), hypertensive disorders (9.2%), and other maternal medical conditions (7.8%).

A higher proportion of stillbirths in non-Hispanic black women, compared with non-Hispanic white women and Hispanic women was associated with obstetric complications (43.5%) and infections (25.2%). This finding combined with the finding that stillbirth in black women often occurs at less than 24 weeks’ gestation suggests that measures aimed at reducing the rate of spontaneous preterm birth in black women could potentially reduce the rate of stillbirth as well (JAMA. 2011 Dec 14;306[22]:2459-68).

Work-up and prevention

Prevention of stillbirth requires that we identify the women at highest risk, and thus far this ability still eludes us. Apart from occurrence of previous stillbirth or pregnancy loss, other risk factors have had limited predictive value in the SCRN analyses and other research.

Biomarkers such as a low PAPP-A during the first trimester and a high AFP in the second trimester – as well as Doppler imaging of the uterine artery – have also been associated with stillbirth, but again, the positive predictive value has been shown to be low (Clin Obstet Gynecol. 2010 Sep;53[3]:597-606). More research is needed to determine if some combination of biochemical markers, imaging, and other risk factors can predict which women are at highest risk.

In the meantime, attention can be paid – in the preconception period if possible – to modifiable risk factors such as maternal obesity, diabetes, and smoking. About 10% of stillbirths are associated with maternal conditions such as hypertension and diabetes, and late stillbirths in particular (28 weeks or later) are associated with maternal medical conditions that are potentially preventable.

Normalization of prepregnancy weight should be a goal, since the overall risk of stillbirth appears to increase independently with increasing BMI. Glycemic control should also be achieved: A recent meta-analysis of preconception and prenatal care of diabetic women estimated “conservatively” that 10% of diabetes-associated stillbirths could be prevented with early detection and glycemic control (BMC Public Health. 2011;11 Suppl 3:S2). Research has also shown that women who quit smoking between their first and second pregnancy reduce their stillbirth risk to that of nonsmokers in the second pregnancy (BJOG. 2007 Jun;114[6]:699-704).

When stillbirth happens, a thorough work-up is recommended in order to counsel for future pregnancies and decrease the risk of recurrence. Evaluations for causes of stillbirth are too often incomplete in the United States for various reasons, including emotional, cultural, and resource factors. Even if a cause is not found, many families appreciate knowing that every effort has been made to determine a cause of death.

Four components of evaluation – autopsy, placental examination, karyotype or microarray analysis, and fetal-maternal hemorrhage testing – have proven to be high-yield tests when performed in all cases of stillbirth.

In the SCRN study, of 512 stillbirths undergoing a complete evaluation, 66.4% had a positive result – defined as abnormalities contributing to a probable or possible cause – for at least one of the first three tests (JAMA. 2011 Dec 14;306[22]:2459-68).

A Dutch study of 1,025 stillbirths similarly demonstrated that all four tests are justified. A test was defined as valuable in this study if it established or excluded a cause of stillbirth. Placental examination was determined to be the most valuable test, helping to determine a cause of death in 95.7% of cases. Autopsy was valuable 72.6% of the time, and cytogenetic analysis was valuable in 29% of cases.

Kleihauer-Betke testing for fetal-maternal hemorrhage was positive in 11.9% of women. However, fetal maternal hemorrhage was considered the cause of death in only 1.3%.of cases because, beyond a positive Kleihauer-Betke test, evidence of fetal anemia confirmed by placental examination and/or autopsy was required for hemorrhage to be considered the cause of death (Am. J. Obstet. Gynecol. 2012;206:53.e1-12). Because Kleihauer-Betke testing is ideally performed before induction, authors of both the SCRN study and the Dutch study believe it is a valuable test to be offered in all cases.

In both studies, the yield of other stillbirth diagnostic tests (for example, maternal serology, hormone assessment, and toxicology screen) was low, indicating that these tests are considered sequential and can be performed only when the clinical history or findings of the four core tests raise suspicion of particular potential causes. Antinuclear antibody testing and TORCH (toxoplasmosis, rubella, cytomegalovirus, herpes simplex) titers have an extremely low yield and are generally not useful.

For detecting genetic abnormalities after stillbirth, it appears that microarray analysis is superior to karyotype analysis. In a SCRN analysis of samples from 532 stillbirths, microarray yielded results more often and identified more genetic abnormalities. Unlike karyotype, it does not require live cells, which makes it preferable for stillbirth evaluation (N Engl J Med. 2012 Dec 6;367[23]:2185-93).

Current research

One of the more significant studies underway on prevention is looking at labor induction as an intervention for reducing stillbirths and improving other perinatal outcomes. The ARRIVE trial (“A Randomized Trial of Induction Versus Expectant Management”), currently in the recruitment stage, will examine outcomes after induction at 39 weeks’ gestation, compared with expectant management in 6,000 patients (clinicaltrials.gov/ct2/show/NCT01990612).

Common wisdom informed by retrospective cohort studies has long told us that inducing labor prior to 41 weeks’ gestation is associated with a higher risk of cesarean delivery in nulliparous women. However, recent observational data have suggested that women whose labor is induced actually have fewer cesarean deliveries and better perinatal outcomes, including a lower risk of stillbirth (AJOG 2012;207:502.e1-8).

In addition, a meta-analysis published in 2014, as the ARRIVE trial was taking shape, reported a 12% reduction in cesarean delivery, and a reduced risk of stillbirth, among women whose labor was induced. The initial cervical score did not impact the main findings (CMAJ. 2014 Jun 10;186[9]:665-73). If these findings are confirmed in the ARRIVE trial, we could see a new opportunity for stillbirth prevention.

Another ongoing study of 10,000 singleton pregnancies – the Nulliparous Pregnancy Outcomes: Monitoring Mothers-to-Be (nuMoM2b) study – may also lead to prevention strategies in women for whom the current pregnancy will lead to their first delivery. Among the questions being examined in this eight-site study are whether sleep-disordered breathing, or apnea, and a supine sleep position are risk factors for adverse pregnancy outcomes including stillbirth.

Supine sleeping in the last month of pregnancy was strongly associated with stillbirth in a recent analysis from the Sydney Stillbirth Study (Obstet Gynecol. 2015 Feb;125[2]:347-55), and an early analysis of a nuMoM2b subset has shown associations between sleep-disordered breathing in midpregnancy and the development of hypertensive disorders of pregnancy, and between sleep-disordered breathing in early- and mid-pregnancy and gestational diabetes (Am J Obstet Gynecol. 2015;212:S424-425).

The possible role of low-dose aspirin in preventing stillbirth also needs more exploration. A recent randomized trial of women attempting to become pregnant after having had one or two prior pregnancy losses found no difference overall in live birth rates between those who took low-dose aspirin and those assigned to placebo. However, there was one subgroup – women with a single loss at less than 20 weeks’ gestation during the previous year – in which live birth rates were higher in the aspirin group (Lancet. 2014 Jul 5;384[9937]:29-36). More research is necessary to determine if low-dose aspirin administration in women with a previous stillbirth improves pregnancy outcome.

Dr. Reddy is a member at the Pregnancy and Perinatology Branch of the Eunice Kennedy Shriver National Institute of Child Health and Human Development. She is a board-certified ob.gyn. and maternal-fetal medicine specialist. She is the program scientist for the Maternal-Fetal Medicine Units Network and for the Stillbirth Collaborative Research Network. The comments and views of the author do not necessarily represent the views of the NICHD.

Stillbirth is a major public health problem, occurring in approximately 1 of every 160 pregnancies in the United States. The rate has remained stagnant since 2006. Prior to that time, from 1990 to 2006, the rate declined somewhat, but only half as much as the decline in infant mortality during this time period. Racial disparities also have persisted, with non-Hispanic black women having more than a twofold increase in risk (Natl Vital Stat Rep. 2012;60:1-22).

Research conducted by the Stillbirth Collaborative Research Network (SCRN) and others has provided us with insight on risk factors and on probable and possible causes of death among stillbirths, which are defined as fetal deaths at 20 or more weeks’ gestation. We know from SCRN data, for instance, that black women are more likely to have stillbirths associated with obstetric complications and infections than white and Hispanic women. However, we still cannot explain a substantial proportion of stillbirths, despite a complete evaluation, or predict who will have a stillbirth.

What we can do as obstetricians is be aware that stillbirth is one of the most common adverse pregnancy outcomes in the United States and counsel women regarding risk factors that are modifiable. Moreover, when stillbirth happens, a complete postmortem evaluation that includes autopsy, placental pathology, karyotype or microarray analysis, and fetal-maternal hemorrhage testing is recommended (Obstet Gynecol. 2009;113[3]:748-61). Recent data show that each of these four components is valuable and should be considered the basic work-up for stillbirth.

Risks and causes

Pregnancy history was the strongest baseline risk factor for stillbirth in an analysis of 614 stillbirths and 1,816 live births in the SCRN’s population-based, case-control study conducted between 2006 and 2008. The SCRN was initiated by the Eunice Kennedy Shriver National Institute of Child Health and Human Development in 2003. This critical population-based study was conducted at 59 U.S. tertiary care and community hospitals in five catchment areas and has been analyzed in more than 15 published reports.

Women with a previous stillbirth have been known to be at 5- to 10-fold increased risk of a recurrence of stillbirth, and the SCRN findings confirmed this. The study added to our knowledge, however, with the finding that even a prior pregnancy loss at less than 20 weeks’ gestation increased the risk for stillbirth.

Other risk factors identified in the study, in addition to race, included having a multifetal pregnancy (adjusted odds ratio of 4.59), diabetes (AOR of 2.50), maternal age of 40 years or older (AOR of 2.41), maternal AB blood type (AOR of 1.96, compared with type O), a history of drug addiction (AOR of 2.08), smoking during the 3 months prior to pregnancy (AOR of 1.55-1.57, depending on amount), and being unmarried and not cohabitating (AOR of 1.69). Regarding racial disparity, the study showed that elevated risk of stillbirth for non-Hispanic blacks occurred predominantly prior to 24 weeks of gestation.

As in prior research, overweight and obesity also conferred elevated risks in the SCRN study (AORs of 1.43 and 1.72, respectively), and these risks were not explained by either diabetes or hypertension (JAMA. 2011;306:2469-79).

The use of assisted reproductive technology was not included in the study’s multivariate model, but previous research has shown a fourfold increased risk of stillbirth for singleton IVF/ICSI pregnancies. The reason is unclear, but the risk appears to be more related to IVF/ICSI rather than the underlying infertility (Hum Reprod. 2010 May;25[5]:1312-6).

A previous preterm or small-for-gestational-age birth has also been shown in prior research to be a significant risk factor for stillbirth. Less clear is the role of previous cesarean delivery in stillbirth risk. An association has been demonstrated in several studies, however, including one involving about 180,000 singleton pregnancies of 23 or more weeks’ gestation. Women in this cohort who had a previous cesarean delivery had a 1.3-fold increased risk of antepartum stillbirth, after controlling for important factors such as race, body mass index (BMI), and maternal disease (Obstet Gynecol. 2010 Nov;116[5]:1119-26).

In another analysis of the SCRN study looking specifically at causes of stillbirth, a “probable” cause of death was found in 61% of cases and a “possible or probable” cause of death in more than 76% of cases. The most common causes were obstetric complications (29.3%), placental abnormalities (23.6%), fetal genetic/structural abnormalities (13.7%), infection (12.9%), umbilical cord abnormalities (10.4%), hypertensive disorders (9.2%), and other maternal medical conditions (7.8%).

A higher proportion of stillbirths in non-Hispanic black women, compared with non-Hispanic white women and Hispanic women was associated with obstetric complications (43.5%) and infections (25.2%). This finding combined with the finding that stillbirth in black women often occurs at less than 24 weeks’ gestation suggests that measures aimed at reducing the rate of spontaneous preterm birth in black women could potentially reduce the rate of stillbirth as well (JAMA. 2011 Dec 14;306[22]:2459-68).

Work-up and prevention

Prevention of stillbirth requires that we identify the women at highest risk, and thus far this ability still eludes us. Apart from occurrence of previous stillbirth or pregnancy loss, other risk factors have had limited predictive value in the SCRN analyses and other research.

Biomarkers such as a low PAPP-A during the first trimester and a high AFP in the second trimester – as well as Doppler imaging of the uterine artery – have also been associated with stillbirth, but again, the positive predictive value has been shown to be low (Clin Obstet Gynecol. 2010 Sep;53[3]:597-606). More research is needed to determine if some combination of biochemical markers, imaging, and other risk factors can predict which women are at highest risk.

In the meantime, attention can be paid – in the preconception period if possible – to modifiable risk factors such as maternal obesity, diabetes, and smoking. About 10% of stillbirths are associated with maternal conditions such as hypertension and diabetes, and late stillbirths in particular (28 weeks or later) are associated with maternal medical conditions that are potentially preventable.

Normalization of prepregnancy weight should be a goal, since the overall risk of stillbirth appears to increase independently with increasing BMI. Glycemic control should also be achieved: A recent meta-analysis of preconception and prenatal care of diabetic women estimated “conservatively” that 10% of diabetes-associated stillbirths could be prevented with early detection and glycemic control (BMC Public Health. 2011;11 Suppl 3:S2). Research has also shown that women who quit smoking between their first and second pregnancy reduce their stillbirth risk to that of nonsmokers in the second pregnancy (BJOG. 2007 Jun;114[6]:699-704).

When stillbirth happens, a thorough work-up is recommended in order to counsel for future pregnancies and decrease the risk of recurrence. Evaluations for causes of stillbirth are too often incomplete in the United States for various reasons, including emotional, cultural, and resource factors. Even if a cause is not found, many families appreciate knowing that every effort has been made to determine a cause of death.

Four components of evaluation – autopsy, placental examination, karyotype or microarray analysis, and fetal-maternal hemorrhage testing – have proven to be high-yield tests when performed in all cases of stillbirth.

In the SCRN study, of 512 stillbirths undergoing a complete evaluation, 66.4% had a positive result – defined as abnormalities contributing to a probable or possible cause – for at least one of the first three tests (JAMA. 2011 Dec 14;306[22]:2459-68).

A Dutch study of 1,025 stillbirths similarly demonstrated that all four tests are justified. A test was defined as valuable in this study if it established or excluded a cause of stillbirth. Placental examination was determined to be the most valuable test, helping to determine a cause of death in 95.7% of cases. Autopsy was valuable 72.6% of the time, and cytogenetic analysis was valuable in 29% of cases.

Kleihauer-Betke testing for fetal-maternal hemorrhage was positive in 11.9% of women. However, fetal maternal hemorrhage was considered the cause of death in only 1.3%.of cases because, beyond a positive Kleihauer-Betke test, evidence of fetal anemia confirmed by placental examination and/or autopsy was required for hemorrhage to be considered the cause of death (Am. J. Obstet. Gynecol. 2012;206:53.e1-12). Because Kleihauer-Betke testing is ideally performed before induction, authors of both the SCRN study and the Dutch study believe it is a valuable test to be offered in all cases.

In both studies, the yield of other stillbirth diagnostic tests (for example, maternal serology, hormone assessment, and toxicology screen) was low, indicating that these tests are considered sequential and can be performed only when the clinical history or findings of the four core tests raise suspicion of particular potential causes. Antinuclear antibody testing and TORCH (toxoplasmosis, rubella, cytomegalovirus, herpes simplex) titers have an extremely low yield and are generally not useful.

For detecting genetic abnormalities after stillbirth, it appears that microarray analysis is superior to karyotype analysis. In a SCRN analysis of samples from 532 stillbirths, microarray yielded results more often and identified more genetic abnormalities. Unlike karyotype, it does not require live cells, which makes it preferable for stillbirth evaluation (N Engl J Med. 2012 Dec 6;367[23]:2185-93).

Current research

One of the more significant studies underway on prevention is looking at labor induction as an intervention for reducing stillbirths and improving other perinatal outcomes. The ARRIVE trial (“A Randomized Trial of Induction Versus Expectant Management”), currently in the recruitment stage, will examine outcomes after induction at 39 weeks’ gestation, compared with expectant management in 6,000 patients (clinicaltrials.gov/ct2/show/NCT01990612).

Common wisdom informed by retrospective cohort studies has long told us that inducing labor prior to 41 weeks’ gestation is associated with a higher risk of cesarean delivery in nulliparous women. However, recent observational data have suggested that women whose labor is induced actually have fewer cesarean deliveries and better perinatal outcomes, including a lower risk of stillbirth (AJOG 2012;207:502.e1-8).

In addition, a meta-analysis published in 2014, as the ARRIVE trial was taking shape, reported a 12% reduction in cesarean delivery, and a reduced risk of stillbirth, among women whose labor was induced. The initial cervical score did not impact the main findings (CMAJ. 2014 Jun 10;186[9]:665-73). If these findings are confirmed in the ARRIVE trial, we could see a new opportunity for stillbirth prevention.

Another ongoing study of 10,000 singleton pregnancies – the Nulliparous Pregnancy Outcomes: Monitoring Mothers-to-Be (nuMoM2b) study – may also lead to prevention strategies in women for whom the current pregnancy will lead to their first delivery. Among the questions being examined in this eight-site study are whether sleep-disordered breathing, or apnea, and a supine sleep position are risk factors for adverse pregnancy outcomes including stillbirth.

Supine sleeping in the last month of pregnancy was strongly associated with stillbirth in a recent analysis from the Sydney Stillbirth Study (Obstet Gynecol. 2015 Feb;125[2]:347-55), and an early analysis of a nuMoM2b subset has shown associations between sleep-disordered breathing in midpregnancy and the development of hypertensive disorders of pregnancy, and between sleep-disordered breathing in early- and mid-pregnancy and gestational diabetes (Am J Obstet Gynecol. 2015;212:S424-425).

The possible role of low-dose aspirin in preventing stillbirth also needs more exploration. A recent randomized trial of women attempting to become pregnant after having had one or two prior pregnancy losses found no difference overall in live birth rates between those who took low-dose aspirin and those assigned to placebo. However, there was one subgroup – women with a single loss at less than 20 weeks’ gestation during the previous year – in which live birth rates were higher in the aspirin group (Lancet. 2014 Jul 5;384[9937]:29-36). More research is necessary to determine if low-dose aspirin administration in women with a previous stillbirth improves pregnancy outcome.

Dr. Reddy is a member at the Pregnancy and Perinatology Branch of the Eunice Kennedy Shriver National Institute of Child Health and Human Development. She is a board-certified ob.gyn. and maternal-fetal medicine specialist. She is the program scientist for the Maternal-Fetal Medicine Units Network and for the Stillbirth Collaborative Research Network. The comments and views of the author do not necessarily represent the views of the NICHD.