Antepartum Fetal Evaluation




Key Abbreviations


American College of Obstetricians and Gynecologists ACOG


Amniotic fluid index AFI


Antiphospholipid antibody syndrome APLAS


Assisted reproductive technology ART


Biophysical profile BPP


Body mass index BMI


Central nervous system CNS


Contraction stress test CST


Deepest vertical pocket DVP


Fetal breathing movement FBM


Fetal movement counting FMC


Human chorionic gonadotropin hCG


Intrauterine growth restriction IUGR


Lecithin/sphingomyelin ratio L/S ratio


Modified biophysical profile mBPP


Multiples of the median MoM


National Center for Health Statistics NCHS


National Institute for Child Health and Human Development NICHD


Nonstress test NST


Perinatal mortality rate PMR


Phosphatidylglycerol PG


Pregnancy-associated plasma protein A PAPP-A


Rapid eye movement REM


Respiratory distress syndrome RDS


Systemic lupus erythematosus SLE


Vibroacoustic stimulation VAS


World Health Organization WHO


Antepartum fetal evaluation is an ever growing and changing science. The goal of evidence-based antepartum fetal evaluation is to decrease perinatal mortality and permanent neurologic injury through judicious use of reliable and valid methods of fetal assessment without acting prematurely to modify an otherwise-healthy pregnancy or providing a false sense of well-being in cases of impending morbidity. The opportunity for obstetric care providers to participate in this delicate balance has been made possible by continued advances in our ability to assess the physiologic well-being of the fetus, concurrent with great improvement in neonatal care and survival. However, despite these advances in antepartum fetal surveillance and the widespread use of antepartum testing programs, the ability of these techniques to prevent intrauterine injury or death remains unproved in many cases. The focus of this chapter is on antepartum evaluation in the United States and similarly technologically advanced and resource-rich countries, noting that the worldwide problem of stillbirth is a vast and compelling area of international interest.




Defining the Problem of Perinatal Mortality


Identification of fetuses at risk for perinatal mortality has historically been the goal of antepartum fetal assessment. Our emerging understanding that long-term neurologic disability is an integrally related and often competing entity to perinatal mortality makes this goal more complex.


The National Center for Health Statistics (NCHS) provides two different definitions for perinatal mortality, acknowledging that variation in definitions and reporting rates both among states in the United States and throughout different countries worldwide makes comparisons difficult; an agenda to develop a classification consensus has been the focus of a number of international committees, including the National Institute for Child Health and Human Development (NICHD). The NCHS National Vital Statistics Report (NVSR) on fetal and perinatal mortality describes two different definitions for perinatal mortality rate (PMR). Definition I includes deaths of infants of less than 7 days of age and fetal deaths of 28 weeks of gestation or more per 1000 live births plus fetal deaths, whereas definition II is more comprehensive and includes infant deaths of less than 28 days of age and fetal deaths of 20 weeks or more per the same denominator. The definitions of PMR provided by the World Health Organization (WHO) and the American College of Obstetricians and Gynecologists (ACOG) differ slightly and include the number of fetuses and live births weighing at least 500 g rather than using a gestational age cutoff. According to the NCHS, “Fetal death means death prior to the complete expulsion or extraction from its mother of a product of human conception, irrespective of the duration of pregnancy and which is not an induced termination of pregnancy. The death is indicated by the fact that after such expulsion or extraction, the fetus does not breathe or show any other evidence of life such as beating of the heart, pulsation of the umbilical cord, or definite movement of voluntary muscles.” The term fetal death is used in these definitions and hereafter in this chapter rather than stillbirth, spontaneous abortion, or miscarriage .


In 2006, about 26,000 fetal deaths occurred in the United States. Although the PMR has fallen steadily in the United States since 1965, the number of fetal deaths has not changed substantially in the past decade ( Fig. 11-1 ). Using NCHS definition I, the PMR reported in 2006 was 6.5 per 1000, and fetal deaths accounted for about 50% of all perinatal mortality in the United States. The PMR varies greatly by maternal race and ethnicity ( Fig. 11-2 ). In 2006, rates (per 1000) were lowest for Asian/Pacific Islander women (4.83), followed by non-Hispanic white (5.34), Hispanic (5.76), and Native American/Alaskan Native women (6.72). The rate for non-Hispanic black women (11.76) was the highest among the racial and ethnic groups and was more than twice the rate for non-Hispanic white women. The significantly greater PMR in blacks results from higher rates of both neonatal and fetal deaths.




FIG 11-1


U.S. trends in fetal mortality rates over time by period of gestation, 1990 through 2006.

(Data from the Centers for Disease Control and Prevention/National Center for Health Statistics, National Vital Statistics System, August 2012.)



FIG 11-2


U.S. fetal mortality rates by maternal race and ethnicity, 2006.

(Data from the Centers for Disease Control and Prevention/National Center for Health Statistics, National Vital Statistics System, August 2012.)


Characteristics of Fetal Death


Another way to consider the contribution of fetal events on PMR is to look at the infant mortality rate ( Fig. 11-3 ). Although the infant mortality rate includes all deaths of infants younger than 1 year of age, 50% of all infant deaths occur in the first week of life, and 50% of these losses result during the first day of life. The infant mortality rate has fallen progressively and even more steeply over time than the fetal death rate, from 47 per 1000 in 1940 to 6.14 per 1000 in 2010. In 2010 about 24,500 infant deaths (6.14 per 1000 live births) were reported, which included 16,200 neonatal deaths (4.05 per 1000 live births), and 12,900 of these were in the first week of life. In 2010, the leading causes of infant mortality were congenital malformations, deformations, and chromosome abnormalities (21%); disorders related to short gestation and low birthweight, not elsewhere classified (17%); sudden infant death syndrome (8%); and maternal complications of pregnancy (6%). Together, these leading causes accounted for over 50% of all infant deaths in the United States in 2010. Clearly, perinatal events play an important role in infant mortality.




FIG 11-3


Relative magnitude of components of fetal and infant mortality, United States 2006.

(Data from the Centers for Disease Control and Prevention/National Center for Health Statistics, National Vital Statistics System, August 2012.)


Causes of Fetal Death


In addition to declining frequency in PMR over time, the overall pattern of perinatal deaths in the United States has changed considerably during the past 40 years. Manning and associates suggest that antepartum deaths may be divided into four broad categories: (1) chronic asphyxia of diverse origin; (2) congenital malformations; (3) superimposed complications of pregnancy, such as Rh isoimmunization, placental abruption, and fetal infection; and (4) deaths of unexplained cause. Fretts and colleagues analyzed the causes of deaths, confirmed by autopsy, in fetuses weighing more than 500 g in 94,346 births at the Royal Victoria Hospital in Montreal from 1961 to 1993. The population studied was predominantly white, participated in prenatal care, and included patients from all socioeconomic groups. Overall, the fetal death rate in this group declined by 70%, from 11.5 per 1000 in the 1960s to 3.2 per 1000 during 1990 to 1993. The decline in the fetal death rate in this cohort was attributed to the prevention of Rh sensitization, antepartum and intrapartum fetal surveillance, improved detection of intrauterine growth restriction (IUGR) and fetal anomalies with ultrasound, and improved care of maternal diabetes mellitus and preeclampsia. The role of antenatal diagnosis and management of congenital malformations and aneuploidy is obviously critical to a goal of reducing perinatal morbidity and mortality and will be discussed separately (see Chapter 9 , Chapter 10 , Chapter 27 ).


Fretts and colleagues noted that most of the deaths in the Canadian cohort occurred between 28 and 36 weeks’ gestation and that the diagnosis of IUGR was rarely identified before death. In addition to IUGR, leading causes of fetal death after 28 weeks’ gestation included abruptio placentae and unexplained antepartum losses. Despite a marked fall in unexplained fetal deaths, from 38.1 to 13.6 per 1000, this category was used for more than 25% of all stillbirths. Fetal to maternal hemorrhage may occur in 10% to 15% of cases of unexplained fetal deaths. Fetal deaths caused by infection, most often associated with premature rupture of the membranes (PROM) before 28 weeks’ gestation, did not decline over the 30 years of the study and accounted for about 19% of fetal deaths. Further population-based analyses of the causes of fetal death have confirmed these findings, including a 2011 evaluation by the Stillbirth Collaborative Research Network, which utilized rigorous methods to classify as many “unexplained” fetal deaths as possible.


In summary, based on available data, about 30% of antepartum fetal deaths may be attributed to asphyxia (IUGR, prolonged gestation), 30% to maternal complications (placental abruption, hypertension, preeclampsia, and diabetes mellitus), 15% to congenital malformations and chromosome abnormalities, and 5% to infection. At least 20% of fetal deaths have no obvious fetal, placental, maternal, or obstetric etiology, and this percentage increases with advancing gestational age. Late-gestation stillbirths are more likely to have no identifiable etiology. The ability of our current methods of surveillance to make an impact on the perinatal mortality will depend on the ability of available tests to predict and predate injury and on use of obstetric interventions to prevent adverse outcomes. In one British series, obstetric and pediatric assessors reviewed the circumstances surrounding each case of perinatal death to identify any avoidable factors that may have contributed to the death. Of the 309 perinatal deaths in this population (half fetal and half in the first week of life), 59% were considered to have had avoidable factors, including 74% of normal-birthweight infants with no fetal abnormalities and no maternal complications. Most avoidable factors were found to be obstetric rather than pediatric or maternal and social. The failure to respond appropriately to abnormalities during pregnancy and labor—including results from the monitoring of fetal growth or intrapartum fetal well-being, significant maternal weight loss, or reported reductions in fetal movement—constituted the largest groups of avoidable factors. This characterization of avoidable factors that contribute to perinatal death has been confirmed in additional studies.


Timing of Fetal Death


Another way to classify fetal deaths may be to differentiate those that occur during the antepartum period and those that occur during labor, or intrapartum deaths . Antepartum fetal death is much more common than intrapartum fetal death, and unexplained fetal death occurs far more commonly than unexplained infant death. In a population-based study in the United States in 2007, the antepartum fetal death rate was 3.7 per 1000, compared with 0.6 per 1000 intrapartum fetal deaths. Although most fetal deaths occur before 32 weeks’ gestation, in planning a strategy for antepartum fetal monitoring, the risk for fetal death must be examined in the population of women who are still pregnant at that point in pregnancy. When this approach is taken, the data would suggest that fetuses at 40 to 41 weeks are at a threefold greater risk and those at 42 or more weeks are at a twelvefold greater risk for intrauterine death than fetuses at 28 to 31 weeks. The risks are even higher in multiple gestations as pregnancy progresses. For twin gestations, the optimal time for delivery to prevent late-gestation perinatal deaths is by 39 weeks, and for triplets, 36 weeks. The issue of timing is also illustrated by a recent cohort study of over 75,000 singleton pregnancies with fetal growth restriction, the focus of which was to find the point at which the competing risks of fetal death and neonatal death were in balance, in order to inform delivery decisions. In this cohort, the balance point was 32 to 34 weeks.


Identifying Those at Risk


Some risk factors have a clear etiologic relationship to fetal compromise and death, such as exposures to teratogens or maternal conditions that alter the fetal environment or blood supply or content. Other risk factors—such as epidemiologic factors that include maternal age, race, and body habitus—have a perhaps more complex and less well-understood link to fetal death risk ( Fig. 11-4 ). Common risk factors for fetal death in the United States are listed in Table 11-1 . Many of these conditions can coexist in individual patients, which makes assessment of the contribution of each factor to perinatal mortality a challenge. A recent concept of looking at individual risk factors as part of a “triple risk model” may prove useful in making sense of this challenge, similar to that used to understand contributors to sudden infant death syndrome (SIDS). In this model, proposed by Warland and Mitchell, an interplay exists among maternal, fetal, and placental factors and a stressor. They posit that whereas these factors in isolation may be insufficient to cause fetal death, they may prove lethal in combination ( Fig. 11-5 ). Also necessary to consider is the contribution of these conditions to fetal injury that results in liveborn children with permanent neurologic compromise; this has yet to be determined but is an important alternative outcome to perinatal mortality that deserves further study.




FIG 11-4


Continuum of certainty in pathophysiology of cause of fetal death. Progressing from left to right on the continuum, levels of certainty increase as to the role of the pathophysiology of a particular condition in causing the fetal death. ALT, alanine aminotransferase; GA, gestational age; SLE, systemic lupus erythematosus.

(Courtesy Professor Gordon Smith. Modified from Reddy UM, Goldenberg R, Silver R, et al. Stillbirth classification: developing an international consensus for research. Executive summary of a National Institute of Child Health and Human Development workshop. Stillbirth Classification of Cause of Death. Obstet Gynecol. 2009;114:901-914.)


TABLE 11-1

COMMON RISK FACTORS FOR FETAL DEATH IN THE UNITED STATES












































































































































































RISK FACTOR PREVALENCE (%) ODDS RATIO
All pregnancies 1.0
Low-risk pregnancies 80 0.86
Obesity:
BMI 25-29.9 21-24 1.4-2.7
BMI >30 20-34 2.1-2.8
Nulliparity compared with second pregnancy 40 1.2-1.6
Fourth child or greater compared with second 11 2.2-2.3
Maternal age (reference: <35 yr):
35-39 yr 15-18 1.8-2.2
≥40 yr 2 1.8-3.3
Multiple gestation:
Twins 2.7 1.0-2.2
Triplets or greater 0.14 2.8-3.7
Oligohydramnios 2 4.5
Assisted reproductive technologies (all) 1-3 1.2-3.0
Abnormal serum markers:
First-trimester PAPP-A <5% 5 2.2-4.0
Two or more second-trimester markers 0.1-2 4.2-9.2
Intrahepatic cholestasis <0.1 1.8-4.4
Renal disease <1 2.2-30
Systemic lupus erythematosus <1 6-20
Smoking 10-20 1.7-3.0
Alcohol use (any) 6-10 1.2-1.7
Illicit drug use 2-4 1.2-3.0
Low education and socioeconomic status 30 2.0-7.0
Fewer than four antenatal visits * 6 2.7
Black (reference: white) 15 2.0-2.2
Hypertension 6-10 1.5-4.4
Diabetes 2-5 1.5-7.0
Large for gestational age (>97% without diabetes) 12 2.4
Fetal growth restriction (%):
<3 3.0 4.8
3-10 7.5 2.8
Previous growth-restricted infant 6.7 2.0-4.6
Previous preterm birth with growth restriction 2 4.0-8.0
Decreased fetal movement 4-8 4.0-12.0
Previous stillbirth 0.5 2.0-10.0
Previous cesarean section 22-25 1.0-1.5
Postterm pregnancy compared with 38-40 wk 2.0-3.0
41 wk 9 1.5
42 wk 5 2.0-3.0

BMI, body mass index; PAPP-A, pregnancy associated plasma protein A.

Modified from Signore C, Freeman RK, Spong CY. Antenatal testing: a reevaluation. Executive Summary of a Eunice Kennedy Shriver National Institute of Child Health and Human Development Workshop. Obstet Gynecol. 2009;113:687-701; and Fretts RC. Stillbirth epidemiology, risk factors, and opportunities for stillbirth prevention. Clin Obstet Gynecol. 2010;53:588-596.

* For stillbirths, 37 weeks’ gestation.




FIG 11-5


Triple risk model in which an interplay of maternal factors, fetal/placental factors, and a stressor contribute to fetal death. Whereas these factors in isolation may be insufficient to cause fetal death, they may prove lethal in combination.

(From Warland J, Mitchell EA. A triple risk model for unexplained late stillbirth. BMC Pregnancy Childbirth. 2014;14:142.)




Details on Select Antenatal Conditions


Maternal Characteristics


Maternal Age


Multiple investigators have found that after controlling for risk factors such as multiple gestation, hypertension, diabetes mellitus, placenta previa and placental abruption, previous abortion, and prior fetal death, women 35 years of age or older have a greater risk for fetal death than women younger than 30 years, and women 40 years or older have an even further increased risk. A J -shaped curve relationship exists between maternal age and fetal deaths, with the highest rates in teenagers and women older than 35 years ( Fig. 11-6 ). The interplay of fetal death, maternal age, and gestational age was demonstrated in a population-based 2006 study in the United States of almost 5.5 million births. In this cohort, compared with their counterparts aged 30 to 34 years at 41 weeks of gestation, women older than 35 to 39 years had the same risk for fetal death at 40 weeks, and women older than 40 years had the same risk at 39 weeks. Only 10% of the women older than 35 years had medical comorbidities, and the results of this study did not change when those women were excluded; this highlights the point that the increase in fetal death risk exists in otherwise healthy older gravidas compared with younger women.




FIG 11-6


Relationship of fetal death and maternal age across gestation.

(From Reddy UM, Ko CW, Willinger M. Maternal age and the risk of stillbirth throughout pregnancy in the United States. Am J Obstet Gynecol. 2006;195:764-770.)


Maternal Race


The variation in fetal death risk in the United States by maternal race is complex, which makes ascertainment of biologic risk factors related to race difficult. Factors that contribute to increased rates of fetal death among black women compared with white women include disparities in socioeconomic status, access to health care, and preexisting medical conditions. A 2009 population-based study of more than 5 million U.S. births demonstrated that the greatest black-and-white disparity is in preterm perinatal death, with a hazard ratio at 20 to 23 weeks of 2.75, which decreases to 1.57 at 39 to 40 weeks. Lower education levels and higher rates of medical, pregnancy, and labor complications contributed more to the adverse outcomes in blacks than in whites, with congenital anomalies more contributory in whites.


Socioeconomic Factors, Prenatal Care, and Substance Abuse


Poor access to prenatal care and poor underlying health and nutrition have been linked to increased risk for fetal death both in developing and developed nations. As with other sociodemographic risk factors, these potential influences on fetal death risk are difficult to quantify and may be additive to other high-risk conditions. Smoking and abuse of alcohol or illicit drugs, along with obesity, represent potentially modifiable risk factors for fetal death. Although these behaviors are attractive candidates for fetal death prevention through counseling and modification of intake, prospective trials of behavior modification strategies have generally been underpowered to detect a difference in fetal death with these interventions.


Maternal Comorbidities


Obesity


Prepregnancy obesity is associated with increased perinatal mortality, especially in late gestation. This has been demonstrated in several large series, including a meta-analysis of 38 studies that included more than 3 million women. The connection between obesity and fetal death is still under investigation and is made more complex by the frequent comorbidities encountered in patients with prepregnancy obesity. Theoretic contributors to adverse perinatal outcomes in this group include placental dysfunction, sleep apnea, metabolic abnormalities, and difficulty in clinical assessment of fetal growth. The strength of the association increases with advancing body mass index (BMI) and with advancing gestational age.


Diabetes Mellitus


Although historically, insulin-dependent diabetes has been a major risk factor for fetal death, the fetal death rate in women with optimal glycemic control now approaches that of women without diabetes. However, the relationship between glycemic control and fetal death remains uncertain. Poor glycemic control is associated with increased perinatal mortality, in large part as a result of congenital anomalies; indicated preterm deliveries; and sudden, unexplained fetal death. A recent population-based study of over 1 million births in Ontario, Canada, revealed an odds ratio (OR) of 2.3 for fetal death among women with pregestational diabetes compared with those without diabetes. No evidence suggests that gestational diabetes controlled by diet alone is associated with increased rates of intrauterine fetal death.


Hypertensive Disorders


Studies have shown conflicting evidence regarding whether fetal death rates with well-controlled preexisting hypertension are comparable to those in the general population or increased. The increased risk for perinatal mortality associated with hypertension is most often related to complicated hyperten­sion, with sequelae of placental insufficiency that include IUGR and oligohydramnios. Proteinuric hypertension, especially preeclampsia with severe features or eclampsia, may be associated with fetal death through placental and coagulation-related pathways, including placental abruption.


Thrombophilia


In general, no demonstrable link has been found between inherited thrombophilia and risk for fetal death. Although initial reports seemed to support an association between fetal death and thrombophilia, such as factor V Leiden mutation and prothrombin gene mutation, large prospective trials have failed to substantiate this association. The presence of circulating maternal antiphospholipid antibodies—in particular lupus antico­agulant, anticardiolipin antibodies, and anti–β 2 -glycoprotein I antibodies—in the antiphospholipid antibody syndrome (APLAS) have been associated with a variety of adverse pregnancy outcomes, including fetal loss. The mechanism of these adverse outcomes remains unclear but likely includes inflammation, thrombosis, and placental infarction. However, the link between fetal death and these antibodies or the presence of APLAS is still under investigation, but evidence is insufficient to conclude that an increased risk exists for fetal death.


Intrahepatic Cholestasis


The cause of fetal death in women with gestational cholestasis remains unknown, and timing and predictive features of impending fetal death remain unpredictable. Fetal deaths in these pregnancies are not preceded by signs of placental insufficiency such as growth restriction or abnormal placental pathology, and normal fetal heart rate tracings proximal to fetal death (i.e., within 24 hours) have often been reported. It has also not been established whether maternal serum levels of bile acids, liver enzymes, or pharmacologic therapy modify or predict the risk of fetal death.


Renal Disease and Systemic Lupus Erythematosus


With chronic maternal renal disease, perinatal outcome is largely associated with the degree of renal dysfunction and the presence of coexisting hypertension or diabetes. Although data are limited by lack of prospective studies with appropriate control groups, the greatest risk for fetal death appears to be in mothers with severe renal impairment (i.e., serum creatinine levels >2.4-2.8 mg/dL). As with maternal renal disease, the prognosis for fetal outcome in women with systemic lupus erythematosus (SLE) is dependent on disease state and comorbid conditions, including hypertension, circulating autoantibodies, and renal involvement. Prognosis for fetal survival in pregnancies complicated by both maternal renal disease and maternal SLE has improved over time with advances in therapies that promote disease quiescence for both of these conditions.


Obstetric Factors


Fertility History and Assisted Reproductive Technology


Multiple aspects of a woman’s obstetric and fertility history may contribute to risk for fetal death in a current pregnancy, including parity, use of assisted reproductive technology (ART), and history of prior adverse obstetric outcomes. Both nulliparity and high parity are associated with an increased risk for fetal death compared with low multiparity (one, two, or three prior births). This association is likely mediated through a variety of sociodemographic risk factors related to overall health and interconception health status, although studies have confirmed the association between parity and fetal death after controlling for several social and medical comorbidities. History of prior adverse pregnancy outcomes—including fetal growth restriction, preterm birth, and fetal death—marks a current pregnancy at risk for fetal death. However, the association with preventable recurrent fetal death is complex and is modified by coexistence of other high-risk conditions. Recurrence risk for fetal death in particular has received recent scrutiny because rates vary dramatically by study population and presence of other risk factors. Given that many fetal deaths occur in pregnancies with no identifiable risk factors and that well-designed studies with appropriate comparison groups (i.e., low-risk women without identifiable risk factors) are lacking, clinician and patient perception of increased risk in a pregnancy subsequent to a fetal death will likely continue to drive management of these patients. Regarding use of ART and risk for fetal death, several systematic reviews have confirmed an independent association between the use of in vitro fertilization in particular and fetal death. However, whether the association is mediated through the technologies themselves or the underlying infertility or through other undetermined mechanisms remains unclear.


Parity


Nulliparity and high parity have both been associated with fetal death in contrast to women having their second child. This association has not been fully explored and may be subject to significant confounding influences, including advanced maternal age, related conditions in nulliparous women with delayed childbearing in developed nations, and other medical and socioeconomic comorbidities in women with high parity.


Multiple Gestations


The higher rate of perinatal mortality in multiple gestations compared with singletons is related both to complications unique to multiple gestations, such as twin-to-twin transfusion syndrome, and to more general complications, such as fetal abnormalities and growth restriction. Additionally, many women who carry more than one fetus have maternal risk factors for increased perinatal mortality, including advanced maternal age and use of ART, and are subject to development of complications such as preeclampsia and preterm delivery. Optimal timing of delivery between 37 and 38 weeks has been considered for twins, compared with 39 to 40 weeks among singletons, because of increased rate of late fetal death in this group. Chorionicity is of paramount importance in determining fetal risk, and rates of adverse outcomes are higher among monochorionic twins.


Early Pregnancy Markers


First- and second-trimester serum markers for aneuploidy, when abnormally low or elevated, have been associated to varying degrees with adverse perinatal outcomes even in the absence of aneuploidy. Biophysical uterine factors have also been studied in this light. Regarding fetal death after 24 weeks, markers of interest include first-trimester levels of pregnancy-associated plasma protein A (PAPP-A) of less than the fifth percentile (0.415 multiples of the median [MoM]); second-trimester free β-human chorionic gonadotropin (free β-hCG), α-fetoprotein (AFP), and inhibin A of more than 2 MoM; and uterine artery pulsatility index above the 90th percentile. The sensitivity and positive predictive value of these markers for fetal death are still under investigation. The pathophysiologic link between these markers and adverse outcomes is unclear and likely variable but most plausibly involves abnormal placental attachment or function.


Amniotic Fluid Abnormalities


The predictive value of either oligohydramnios or polyhydramnios for adverse pregnancy outcomes, in particular fetal death, typically lies in their association with other abnormal conditions, such as maternal diabetes mellitus, hypertensive disorders, rupture of membranes, fetal growth restriction, or fetal anomalies. Isolated oligohydramnios and polyhydramnios have not been conclusively linked to increased risk for fetal death; nevertheless, evaluation of amniotic fluid volume as a marker of long-term fetal health status is a mainstay of antepartum fetal evaluation.


Fetal Growth Restriction


IUGR is a well-known risk factor for perinatal death that has historically been underrecognized before fetal death. Placental dysfunction is commonly implicated in nonmalformed and chromosomally normal IUGR fetuses. This topic is reviewed in further detail in Chapter 33 .


Postterm Pregnancy


The definition of postterm pregnancy has been reevaluated in the past decade based on reappraisal of the peak time of fetal risk in relation to the 40-week mark (see Chapter 36 ). The pathophysiology of increased fetal death risk in the postterm pregnancy is thought to be mediated by impaired placental oxygen exchange and is often associated with oligohydramnios. Traditionally oligohydramnios has been used as a marker for increased risk in the postterm pregnancy for which intervention in the form of delivery is thought to be necessary, although as described previously, whether oligohydramnios is independently associated with fetal death in pregnancies after 40 weeks’ gestation is unproven.


Fetal Malformations


Pregnancies complicated by major fetal anomalies are at increased risk of stillbirth, independent of coexisting fetal growth restriction, as demonstrated recently in a large retrospective cohort study. The overall stillbirth rate among fetuses with a major anomaly was 55 per 1000 compared with 4 per 1000 in nonanomalous fetuses with an adjusted odds ratio of 15. The rate of stillbirth was highest in fetuses with congenital cardiac defects. The authors caution that in this group of at-risk fetuses in particular, “health care practitioners caring for these patients should weigh the competing risks of postnatal mortality with antenatal death.”


Conclusions


The best use of available antenatal testing modalities may vary according to the risk profile of each individual pregnancy. Discussion of condition to specific testing will be undertaken after review of the individual testing modalities described later.




Potential Utility of Antepartum Fetal Testing


Can antepartum fetal deaths and injury be prevented? Before using antepartum fetal testing, the obstetrician must ask several important questions:



  • 1.

    Does the test provide information not already known by the patient’s clinical status?


  • 2.

    Can the information be helpful in managing the patient?


  • 3.

    If an abnormality is detected, is a treatment available for the problem?


  • 4.

    Could an abnormal test result lead to increased risk for the mother or fetus?


  • 5.

    Will the test ultimately decrease perinatal morbidity and mortality?



A large body of clinical and research experience suggests that antepartum fetal assessment can have a significant impact on the frequency and causes of antenatal fetal deaths. However, according to several reviews of the benefits and costs of antenatal testing, “strong evidence for the efficacy of antepartum testing is lacking.” Unfortunately, few of the antepartum tests commonly used in clinical practice today have been subjected to large-scale prospective and randomized evaluations that can speak to the true efficacy of testing. In most cases, the test has been applied and good perinatal outcomes have been observed; therefore the test has gained further acceptance and has been used more widely. In such cases, it is uncertain whether the information provided by the test has accurately led to the improved outcomes or the total program of care has made the difference. When prospective randomized investigations are conducted, large numbers of patients must be studied because many adverse outcomes, such as intrauterine death, are uncommon even in high-risk populations. For example, although several controlled trials have failed to demonstrate improved outcomes with nonstress testing, the study populations ranged from only 300 to 530 subjects.


To determine the clinical application of antepartum diagnostic testing, the predictive value of the tests must be considered. The sensitivity of the test is the probability that the test will be positive or abnormal when the disease is present; the specificity of the test is the probability that the test result will be negative when the disease is not present. Note that the sensitivity and specificity refer not to the actual numbers of patients with a positive or abnormal result but to the proportion or probability of these test results. The predictive value of an abnormal test would be that fraction of patients with an abnormal test result who have the abnormal condition, and the predictive value of a normal test would be the fraction of patients with a normal test result who are normal.


Antepartum fetal tests may be used to screen a large obstetric population to detect fetal disease. In this setting, a test of high sensitivity is preferable to minimize the risk of missing a patient whose fetus might be compromised. It would be prudent to be willing to overdiagnose the problem—that is, to accept some false-positive diagnoses. In further evaluating the patient whose fetus may be at risk, and when attempting to confirm the presence of disease, a test of high specificity is preferable. It is best not to intervene unnecessarily and deliver a fetus that was doing well. In this setting, multiple tests may be helpful. When multiple test results are normal, they tend to exclude disease; however, when all are abnormal, they tend to support the diagnosis of fetal disease.


The prevalence of the abnormal condition has great impact on the predictive value of antenatal fetal tests and the number needed to evaluate with testing and to treat with interventions (delivery) to presumably prevent fetal death. The impact of these parameters on the utility of testing was illustrated in a decision analysis of the risks and benefits of antepartum testing late in pregnancy for women 35 years or older using the McGill Obstetric/Neonatal Database (MOND) to obtain risk estimates. In this model, as in practice, the relative benefit of antenatal testing lies in the balance of the number of fetal deaths prevented with the number and type of interventions required to prevent them. At an estimated risk for unexplained fetal death of 5.2 per 1000 pregnancies (nulliparas ≥35 years of age in the McGill cohort), 863 additional antenatal tests, 71 additional inductions of labor, and 14 additional cesarean deliveries would be required to prevent one additional fetal death using this model. Comparatively, using the same model at an estimated risk of 1 to 2 per 1000 pregnancies, 2862 additional antenatal tests, 233 additional inductions, and 44 additional cesarean deliveries would be required to prevent one additional fetal death. Thus the number needed to evaluate and treat to prevent one fetal death decreases as risk for fetal death increases in the population being tested.


In interpreting the results of studies of antepartum testing, the obstetrician must consider the application of that test to his or her own population. If the study has been done in a population of patients at great risk, it is more likely that an abnormal test will be associated with an abnormal fetus. If the obstetrician is practicing in a community with patients who are, in general, at low risk, an abnormal test result will more likely be associated with a false-positive diagnosis.


For most antepartum diagnostic tests, a cutoff point used to define an abnormal result must be arbitrarily established. The cutoff point is selected to maximize the separation between the normal and diseased populations ( Fig. 11-7 ). Changing the cutoff will have a great impact on the predictive value of the test. For example, suppose that 10 accelerations in 10 minutes were required for a fetus to have a reactive nonstress test (NST; threshold A). The fetus that fulfilled this rigid definition would almost certainly be in good condition. However, many fetuses that failed to achieve 10 accelerations in 10 minutes would also be in good condition but would be judged to be abnormal using this cutoff. In this instance, the test would have many abnormal results; it would be highly sensitive and capture all of the abnormal fetuses, but it would have a low specificity. If the number of accelerations required to pass an NST were lowered to one in 10 min, it would decrease the sensitivity of the test (threshold C). That is, the clinician might miss a truly sick fetus. At the same time, however, the specificity of the test or its ability to predict that percentage of patients who are normal would improve. Using the criterion of two accelerations of the fetal heart rate in 20 min for a reactive NST (threshold B), it is hoped that the test will have both high sensitivity and high specificity.




FIG 11-7


Hypothetical distribution of test results in a normal and diseased population demonstrating the differences in test sensitivity and specificity with a change in test threshold. Making it more difficult for the fetus to pass the test by raising the test threshold ( A ) will increase the sensitivity but decrease the specificity of the test. On the other hand, making the test easier to pass by decreasing the test threshold ( C ) will increase the specificity of the test but decrease the sensitivity.

(Modified from Carpenter M, Coustan D. Criteria for screening tests for gestational diabetes. Am J Obstet Gynecol. 1982;144:768.)




What Do These Tests Tell Us About the Fetus?


Fetal State


To be able to diagnose suspected fetal compromise using tests of fetal biophysical state, blood flow, and heart rate, we must be able to appreciate how these parameters appear under normal conditions and in response to suboptimal conditions.


Regarding fetal biophysical characteristics, it must be appreciated that during the third trimester, the normal fetus can exhibit marked changes in its neurologic state. Four fetal states have been identified. The near-term fetus spends about 25% of its time in a quiet sleep state (state 1F) and 60% to 70% in an active sleep state (state 2F). Active sleep is associated with rapid eye movement (REM). In fetal lambs, electrocortical activity during REM sleep is characterized by low-voltage, high-frequency waves. The fetus exhibits regular breathing movements and intermittent abrupt movements of its head, limbs, and trunk. The fetal heart rate in active sleep (state 2F) exhibits increased variability and frequent accelerations with movement. During quiet, or non-REM, sleep, the fetal heart rate slows, and heart rate variability is reduced. The fetus may make infrequent breathing and startled movements. Electrocortical activity recordings at this time reveal high-voltage, low-frequency waves. Near term, periods of quiet sleep may last 20 minutes, and those of active sleep, about 40 minutes. The mechanisms that control these periods of rest and activity in the fetus are not well established. External factors such as the mother’s activity, her ingestion of drugs, and her nutrition may play a role. Specific factors that may decrease fetal movement in the third trimester include fetal anomalies, particularly central nervous system (CNS) anomalies; maternal exposures, including corticosteroids, sedatives, smok­ing, and anxiety; low amniotic fluid volume; and decreased placental blood flow due to placental insufficiency.


When evaluating fetal condition using the NST or the biophysical profile (BPP), the clinician must consider whether a fetus who is not making breathing movements or showing accelerations of its baseline heart rate is in a quiet sleep state or is neurologically compromised. In such circumstances, prolonging the period of evaluation usually allows a change in fetal state, and more normal parameters of fetal well-being will appear.


Regarding regulation of fetal heart rate and blood flow, fetal adaptation to hypoxemia is mediated through changes in heart rate and redistribution of cardiac output. However, changes in fetal cardiac output are generally observed during hypoxemia only with coexisting acidemia. In response to sudden hypoxemia, fetal heart rate slowing and increased variability can be observed initially through vagally mediated chemoreceptor responses. With prolonged hypoxemia (30-60 min), increasing levels of circulating adrenergic agonists and modulation of vagal activity by endogenous opiates lead to a fetal heart rate return to or rise from the previous baseline. Development of acidemia on top of hypoxemia can accelerate the rate of fetal deterioration and amplify the hypoxemia by a shift of the oxyhemoglobin dissociation curve to the right, which further reduces the oxygen-carrying capacity of fetal blood and eventually leads to a redistribution of cardiac output that can be appreciated as a “brain-sparing” effect in the evaluation of fetal blood flow. Redistribution of blood flow in the compromised fetus preferentially preserves perfusion not only to the brain but also to the heart and adrenal glands.


Fetal movement is a more indirect indicator of fetal oxygen status and CNS function, and decreased fetal movement is noted in response to hypoxemia. However, gestational development of fetal movement must be considered when evaluating fetal well-being as marked by fetal activity. Periods of absent fetal movement become more prolonged as gestation advances, and normal fetuses progressively exhibit longer periods of quiescence as the late second and third trimesters advance. Up to and perhaps longer than 40 minutes of fetal inactivity at 40 weeks may be a normal finding, compared with less than 10 minutes at 20 weeks and less than 20 minutes at 32 weeks. Keeping these trends in mind, an abnormal degree or absence of fetal movement can be an appropriate marker for fetal hypoxemia. However, fetal activity levels have been seen in animal studies to adapt to inducement of hypoxemia with a resumption of fetal breathing and body movements after a prolonged period of hypoxemia, especially if induced gradually. Therefore observation of these fetal states during antenatal testing does not guarantee a normoxic fetus.




Biophysical Techniques of Fetal Evaluation


Descriptions of the predictive value of several commonly performed antenatal tests of fetal well-being are presented in Table 11-2 , and details on each methodology are given in the sections that follow.



TABLE 11-2

COMPARISON OF SELECTED ANTENATAL TESTS
























TEST FALSE-NEGATIVE RATE (%) FALSE-POSITIVE RATE (%)
Contraction stress test 0.04 35-65
Nonstress test 0.2-0.8 55-90
Biophysical profile 0.07-0.08 40-50
Modified biophysical profile 0.08 60

Modified from Signore C, Freeman RK, Spong CY. Antenatal testing: a reevaluation. Executive Summary of a Eunice Kennedy Shriver National Institute of Child Health and Human Development Workshop. Obstet Gynecol. 2009;113:687-701.


Maternal Assessment of Fetal Activity


Studies performed using real-time ultrasonography have demonstrated that during the third trimester, the human fetus spends 10% of its time making gross fetal body movements and that 30 such movements are made each hour. Periods of active fetal body movement last about 40 minutes, whereas quiet periods last about 20 minutes. Patrick and colleagues noted that the longest period without fetal movements in a normal fetus was about 75 minutes. The mother is able to perceive about 70% to 80% of gross fetal movements. The fetus does make fine body movements such as limb flexion and extension, hand grasping, and sucking; these probably reflect more coordinated CNS function, although the mother is generally unable to perceive these fine movements. Fetal movement appears to peak between 9:00 pm and 1:00 am , a time when maternal glucose levels are falling. In a study in which maternal glucose levels were carefully controlled with an artificial pancreas, Holden and coworkers found that hypoglycemia was associated with increased fetal movement. Fetal activity does not increase after meals or after maternal glucose administration, but it may increase during exposure to music, as was demonstrated in one novel randomized trial.


The decrease in fetal movement with hypoxemia makes maternal assessment of fetal activity a potentially simple and widely applicable method of monitoring fetal well-being. However, prospective trials of this method for prevention of perinatal mortality have failed to conclusively show benefit. Neldam demonstrated a 73% reduction in avoidable fetal deaths in a prospective trial of more than 1500 women instructed to count fetal movements. In contrast, a subsequent international trial of more than 68,000 women randomized to routine fetal movement assessment versus no formal assessment of fetal activity failed to show a significant reduction in fetal deaths in the group randomized to routine movement counting. Significant differences were noted in the method of fetal movement counting (FMC), the definition of “abnormal” fetal movements, patient compliance with the intervention, and provider response to patients who presented to care as a result of an abnormal fetal movement count in the latter, compared with the former, trial and throughout the literature on this topic. This illustrates the difficulties in validating and reproducing the results of these trials and the uncertain clinical benefit that may be derived from introducing maternal assessment of fetal movement into routine clinical practice. A 2007 Cochrane review of four trials that involved more than 71,000 women concluded that evidence is insufficient to recommend routine FMC to prevent fetal death.


Despite these results, this type of fetal assessment may offer some advantages. Although the range in fetal activity will be wide but normal, with FMC, each mother and her fetus serve as his or her own control. Factors that affect maternal perception of fetal movement are not well understood. Fetal and placental factors that may contribute include placental location, the length and type of fetal movements, and amniotic fluid volume (AFV), although whether AFV affects maternal perception or actual fetal movement is unclear. Maternal factors that may influence the evaluation of fetal movement include maternal activity, parity, obesity, medications, and psychological factors, including anxiety. Studies of these associations have demonstrated conflicting results. About 80% of all mothers are able to comply with a program of counting fetal activity.


Several methods have been used to monitor fetal activity in research and clinical practice. These methods include FMC over a prescribed time period, such as 30 to 60 minutes one to three times daily, or conversely a target number of fetal movements to be counted over a variable time range. A variety of “normal” and “abnormal” FMC results or thresholds have been proposed, to which the patient should be instructed to respond by presenting for further evaluation of the condition of the fetus ( Fig. 11-8 ). These potential triggers for further evaluation include fewer than three movements in 1 hour or no movements for 12 hours, Sadovsky’s “movement alarm signal”; fewer than three movements an hour for 2 consecutive days; or inability to count 10 movements in a 12-hour period, the Cardiff “count-to-10” method advocated by Pearson and Weaver. The count-to-10 method has received wide scrutiny and is used perhaps most frequently in clinical practice. The use of this technique as a screening tool has most recently been reexamined by Froen and associates. In a cohort of 1200 women instructed to start their fetal movement count at the first convenient time of the day, they found that the mean time to count to 10 was less than 10 minutes, compared with significantly longer and more variable average times reported in previous investigations. Noting the variation in defining normal and abnormal patterns and maternal perception of fetal movement as mentioned previously, Froen and associates concluded that significant research is needed to better define the changes in fetal activity patterns and perception associated with good and adverse perinatal outcomes.


Mar 31, 2019 | Posted by in OBSTETRICS | Comments Off on Antepartum Fetal Evaluation

Full access? Get Clinical Tree

Get Clinical Tree app for offline access