| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Division of Maternal-Fetal Medicine and the Section of Epidemiology and Biostatistics, Department of Obstetrics, Gynecology and Reproductive Sciences, University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School/Saint Peter’s University Hospital, New Brunswick, New Jersey.
Address reprint requests to: John C. Smulian, MD, MPH, UMDNJ-RWJMS/Saint Peter’s University Hospital, 254 Easton Avenue, MOB–4th Floor, New Brunswick, NJ 08903-0591; E-mail: smuliajc{at}umdnj.edu.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: This study examined 10,614,679 non-anomalous singleton pregnancies delivering at or beyond 24 weeks’ gestation, derived from the U.S. linked birth/infant death data sets, 1995–1997. Fetal death rates for pregnancies at low risk were compared with pregnancies complicated by chronic hypertension, gestational hypertensive disorders, diabetes, small for gestational age infants, and abruption. Adjusted relative risks as well as population-attributable risks for fetal death were derived by gestational age for each high-risk condition compared with low-risk pregnancies.
RESULTS: The fetal death rate for low-risk pregnancies was 1.6 per 1000 births. Adjusted relative risk for fetal death was 9.2 (95% confidence interval [CI] 8.8, 9.7) for abruption, 7.0 (95% CI 6.8, 7.2) for small for gestational age infants, 1.4 (95% CI 1.3, 1.5) for gestational hypertensive disorders, 2.7 (95% CI 2.4, 3.0) for chronic hypertension, and 2.5 (95% CI 2.3, 2.7) for diabetes. Fetal death rates were lowest between 38 and 41 weeks. The fetal death rate (per 1000 births) for these high-risk conditions was 61.4, 9.6, 3.5, 7.6, and 3.9, respectively. Almost two thirds of fetal deaths were attributable to the pregnancy complications examined.
CONCLUSION: High-risk conditions in pregnancy are associated with an increased risk for fetal death, particularly in the third trimester. Delivery should be considered at 38 weeks, but no later than 41 weeks, for these pregnancies.
The risk for fetal death is increased in pregnancies complicated by chronic hypertension, hypertensive diseases of pregnancy (pregnancy-induced hypertension, preeclampsia, and eclampsia), diabetes, small for gestational age births, and placental abruption.1,2 The presence of any of these conditions is commonly considered an indication for antenatal fetal testing to assure fetal well-being.3–6 Several different methods for monitoring fetal well-being are used in the United States. These include "kick counts," which require only time and effort from the pregnant woman, as well as technology-dependent methods that include fetal heart rate monitoring, ultrasound examinations with biophysical profile evaluations and amniotic fluid volume assessments, and Doppler flow velocimetry.3 The primary goal of fetal testing by any method is to prevent fetal death. Traditionally, recommendations based on empirical evidence suggest initiating testing around 32 weeks’ gestation in the presence of most high-risk conditions,3,7 but there are no consistent recommendations as to when to discontinue testing and deliver the fetus.
To make decisions about when to initiate and when to discontinue fetal testing, it is important to understand the influence of each high-risk condition on the timing of fetal death. There is little available information, if any, to describe the timing of fetal death in high-risk conditions in comparison with otherwise normal pregnancies, which would allow an appropriate judgment of how to time antepartum fetal testing. Therefore, this study was designed to determine whether the presence of specific high-risk conditions alters either the risk or timing of fetal death in the United States compared with apparently low-risk pregnancies without these conditions, and thereby to judge when to apply fetal surveillance techniques to prevent fetal death.
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Analyses were restricted to singleton births delivered at or beyond 24 completed weeks of gestation. The timing of fetal death was assigned using the gestational age at delivery as a surrogate because the exact timing of fetal death was not available. Gestational age assignment was predominantly based on the date of last menstrual period (LMP). In about 5% of births, the date or month of LMP was missing, and therefore it was imputed.9 Inconsistent gestational age for birth weight was replaced by a clinical estimate in a small fraction of births. The gestational age imputations and the replacement of clinical estimate were performed by the National Center for Health Statistics before public release of data.
The vital statistics data contain information on several sociodemographic factors, as well as medical and obstetric complications of pregnancy. Sociodemographic characteristics that were examined included maternal age, gravidity, marital status (married or unmarried), maternal education (years of schooling), cigarette use (yes or no), maternal race (whites, blacks, and other race/ethnicity), and prenatal care. Complications of pregnancy examined were chosen because of their reported association with increased risk for fetal death and because these high-risk conditions encompass the majority of the common indications for antepartum testing and are risk factors for fetal death. These included hypertensive disorders, diabetes, and placental abruption. All complications of pregnancy are recorded on vital statistics data using a check-box format,10 indicating the presence or absence of the condition. Also included in the analysis were small for gestational age infants, defined as birth weight below the 10th centile for gestational age, derived from singleton births in the United States for 1995–1997 (internally derived data).
Hypertensive diseases were classified as either chronic hypertension, pregnancy-induced hypertension, or eclampsia. For purposes of analysis, the latter two were combined into a category of gestational hypertensive disorders. Chronic hypertension was recorded if the woman had elevated blood pressures before pregnancy, or prior to 20 weeks’ gestation. The diabetes category included both pre-existing and gestational diabetes. Placental abruption was defined as the complete or partial separation of the placenta prior to delivery. A clinical diagnosis of this condition is usually confirmed at delivery. No data were available to determine when high-risk conditions were identified during pregnancy.
To determine the fetal death rate in pregnancies considered low-risk for fetal death, an examination of fetal death was performed for pregnancies in the absence of vaginal bleeding, placenta previa, placental abruption, hypertensive disorders (chronic and gestational hypertension), diabetes, and small for gestational age births. The risk of fetal death (per 1000 births) at each week of gestation (24 to 42+ completed weeks), as well as for each high-risk condition, was calculated using the number of deaths at week "x" divided by the total number of births (fetal death plus live births) at week "x." Adjusted odds ratio and 95% confidence interval for fetal death at each week of gestation were derived from multivariable logistic regression models for each high-risk pregnancy complication, including chronic hypertension, gestational hypertensive disorders, diabetes, small for gestational age births, and abruption using pregnancies at low risk for fetal death as the reference group.
Variables considered for adjustment as potential confounders included maternal age, gravidity, marital status, cigarette use, maternal education, absence of prenatal care, race/ethnicity, anemia, and intrapartum fever. Nonlinear relationship between fetal deaths and maternal age was assessed by including a second-order polynomial for maternal age in the regression models. The specific variables included for adjustment are noted in the Results section. Because odds ratio is not a good approximator of relative risk (RR) when the incidence of the outcome in the presence of a risk factor is high, we transformed the adjusted odds ratios to RRs.11 The population attributable risk (percent) was derived for each high-risk condition to assess the impact of each condition on fetal death.12
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Compared with the low-risk pregnancies, infants of mothers with hypertensive diseases or abruption were delivered earlier and weighed less.
|
. Death rates for the examined complications are higher and roughly parallel to the rates for low-risk pregnancies across the entire spectrum of gestational age. All fetal death rates decrease with advancing gestational age; however, the death rate related to abruption decreases much less dramatically. Overall fetal death rates were lowest from 38 to 41 weeks and increased thereafter for most conditions. The fetal death rates for low-risk pregnancies and the adjusted RRs for fetal death by pregnancy complication are reported for each week of gestation in Table 2. For all pregnancy complications examined, the RR for fetal death was increased across almost all gestational ages. The magnitude of increased risk was greatest for pregnancies complicated by abruption, especially in the third trimester, followed by the presence of a small for gestational age infant. For chronic hypertensive diseases and diabetes, the magnitude of higher RR for fetal death over low-risk pregnancies tended to increase two- to three-fold near 34 weeks. The RR for fetal death for small for gestational age births remained elevated more than four-fold throughout gestation. The proportion of all fetal deaths occurring at 38 weeks and beyond among low-risk pregnancies was 34.4% (4787 of 13,935); the proportion of fetal deaths at or beyond 38 weeks among high-risk conditions was 28.2% (4442 of 15,735). Overall, high-risk pregnancies accounted for 51.9% of the fetal deaths at or beyond 38 weeks.
|
|
. Abruption conferred the overall highest RR, followed by small for gestational age births, chronic hypertension, diabetes, and gestational hypertensive disorders. The population-attributable risks for fetal death by pregnancy complications also are described in Table 3. The attributable risks were adjusted for the other complicating conditions, as well as for gestational age at delivery, to isolate the population risk conferred by each condition. Over 50% of fetal deaths were attributable to either the presence of a small for gestational age birth or abruption. Hypertensive diseases accounted for 8.1% and diabetes for 0.4% of fetal deaths.
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) because many of these pregnancies may have had fetal testing with obstetric intervention for abnormal tests. It is also possible that some pregnancies were incorrectly categorized as low risk because of missing information about high-risk conditions. This would lead to further underestimates of the RR or fetal death. It is likely that many patients with the high-risk conditions may have been managed with antepartum testing, so our observations may not reflect "natural history." The fetal death rates may have been influenced by physicians’ decisions about the timing of testing and timing of interventions. Even though there are no randomized controlled trials of antepartum fetal testing for prevention of fetal death, there are other data that suggest that testing has reduced the fetal death rate in some situations.13 Therefore, the fetal death rates for high-risk conditions reported in this analysis may be biased toward an optimum lower rate. Although our results represent fetal death risks under current management practices in the United States, some additional inferences regarding management may be justified by this analysis.
The consistently increased risk for fetal death in the examined high-risk conditions suggests that antenatal fetal testing is justified in their presence. Decisions as to when to start fetal testing should be based on the magnitude of risk for fetal death. The risks of the false-positive examination also must be considered, along with the potential benefits conferred by an intervention (such as delivery) in the presence of a true abnormal test. For most preterm pregnancies the threshold for intervention (delivery) should be higher than for term pregnancies, because severely premature infants have greater risks for morbidity and mortality. Therefore, at-risk preterm fetuses that lack a specific indication for delivery should be monitored closely. Decisions for when to initiate fetal monitoring also depend on a determination of when the risk for death increases significantly compared with pregnancies at low risk for fetal death. Although a definition of significantly increased risk is somewhat subjective, a two- to three-fold increase in fetal death RR to low-risk pregnancies would seem to be a reasonable threshold for initiating fetal testing. The specific testing modality should be tailored to the indication but could include some combination of kick counts, ultrasound biometry, Doppler velocimetry, fetal biophysical profile assessments, and fetal heart rate monitoring.
Examination of the patterns of derived RRs from this study suggests that some form of fetal monitoring should be in place by 32–34 weeks for women with chronic hypertension or diabetes. Because of the consistently high risk conferred by the presence of small for gestational age, testing should be considered when a small fetus (<10th centile) is diagnosed, especially when there is an abnormal growth velocity or biometry is asymmetric. Based on the information provided from this study, delivery should be considered as early as 38 weeks, but no later than 41 weeks, for all pregnancies with these complications instead of continued surveillance. This is the time when fetal death rates are at or near their lowest levels and yet remain well above the fetal death rates for more normal pregnancies. Although we do not have information from this cohort regarding the use or timing of antepartum testing in patients with high-risk conditions, it still seems reasonable to conclude that the risk for fetal death in high-risk conditions is the lowest at 38–41 weeks.
For women with gestational hypertensive disorders, the RR for fetal death remains below 2 until 39 weeks. This may reflect the modifying effects of intervention by delivery on the risk for death and not the true risks, per se. Certainly, women with the more advanced degrees of gestational hypertensive disorders (preeclampsia and eclampsia) should be considered for delivery at the time of diagnosis, regardless of the fetal condition, because of the maternal risks.
The correct assignment of gestational age is important for calculating age-specific fetal death risks. Because fetal death always occurs prior to delivery, the use of delivery gestational age as a surrogate for age at fetal death will likely overestimate the gestational age assigned for the actual death. This may result in some overestimation of the number of infants that are small for gestational age. Nevertheless, most fetal deaths are detected close to time of death because mothers usually seek care when fetal movements decrease or stop.14 It has been reported that, when managed expectantly, at least 90% of women will spontaneously develop labor within 2 weeks of the diagnosis of fetal demise.14 In recent years, the majority of pregnancies with a demise are not managed expectantly, but are rather induced. Therefore, we believe that gestational age at delivery is a reasonable estimate for gestational age of fetal death in most cases. The vital statistics database does not distinguish between antepartum and intrapartum fetal death, but because of the low rate of intrapartum death with current methods of fetal monitoring during labor,15 the overwhelming majority of fetal deaths recorded will be antepartum.
Of particular interest is the finding that over 50% of fetal deaths are attributable to either the presence of a fetus that is small for gestational age or abruption. This suggests that it is mandatory to develop strategies directed toward risk assessment and management of these two conditions to reduce the number of fetal deaths. Fetuses that are at risk for being small for gestational age due to uteroplacental insufficiency may be identified in the second trimester using ultrasound Doppler velocimetry screening of the uterine arteries.16 They also may be identified from a history of previous growth restriction. Similarly, women who are at high-risk for abruption may be identified by history or the presence of associated at-risk conditions.17–20 Nevertheless, a widely accepted definition of high-risk status for abruption that carries acceptable predictive values has not been established. Optimum testing modalities that would minimize false-positive tests leading to unnecessary interventions also remain to be defined. It should be noted that by excluding cases of unexplained vaginal bleeding, many cases of true abruption were not analyzed. This may have biased the results toward stronger associations between abruption and fetal deaths.
It is important to acknowledge that recommendations for initiating testing to prevent fetal death that are inferred from these population-based data should be interpreted cautiously and may not substantially change current management of high-risk pregnancies in the United States. This is particularly true if most of such pregnancies are already managed using various testing modalities. It may not be possible to significantly improve fetal death rates before term if management is already optimized. Unfortunately, there are no population-based studies that describe the use of fetal testing in high-risk pregnancies that describe variations in use or appropriateness of fetal testing; however, delivery as soon as a patient reaches term, rather than continuation of pregnancy for these conditions, would reduce an important proportion of fetal death. Our data supports a strategy of consideration for delivery beginning around 38 weeks of pregnancy, and certainly by 41 weeks, in the presence of high-risk conditions, because there is little additional fetal benefit to be gained from pregnancy prolongation to offset the persistently increased risk for fetal death. This appears to lend credence to traditional obstetric dogma that women with high-risk conditions generally should be delivered by their due date.
|
Footnotes |
|---|
Received March 4, 2002. Received in revised form May 18, 2002. Accepted May 30, 2002.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Fretts RC, Boyd ME, Usher RH, Usher HA. The changing pattern of fetal death, 1961–1988. Obstet Gynecol 1992;79:35–9.
3. American College of Obstetricians and Gynecology. Antepartum fetal surveillance. 2000 compendium ACOG practice bulletin no. 9. Washington, DC: American College of Obstetricians and Gynecologists, 1999:927–37.
4. Harman CR, Menticoglou SM. Fetal surveillance in diabetic pregnancy. Curr Opin Obstet Gynecol 1997;9: 83–90.[Medline]
5. Snijders R, Hyett J. Fetal testing in intrauterine growth retardation. Curr Opin Obstet Gynecol 1997;9:91–5.[Medline]
6. Sholl JS. Abruptio placenta: Clinical management in non-acute cases. Am J Obstet Gynecol 1987;156:40–51.[Medline]
7. Antepartum fetal monitoring. In: Freeman RK, Garite TJ, Nageotte MP, eds. Fetal heart rate monitoring, 2nd ed. Baltimore: Williams & Wilkins, 1991:156–78.
8. MacDorman MF, Atkinson JO. Infant mortality statistics from the linked birth/infant death data set–1995 period data. Natl Vital Stat Rep 1998;4 Suppl 2:1–220.
9. Taffel S, Johnson D, Heuse R. A method of imputing length of gestation on birth certificates. Vital Health Stat 2 1982;93:1–11.
10. Taffel SM, Ventura SJ, Gay GA. Revised U.S. certificate of birth—New opportunities for research on birth outcome. Birth 1989;16:188–93.[Medline]
11. Zhang J, Yu FK. What’s the relative risk? A method of correcting the odds ratio in cohort studies of common outcomes. JAMA 1998;280:1690–1.
12. Eide GE, Gefeller O. Sequential and average attributable fractions as aids in the selection of preventive strategies. J Clin Epidemiol 1995;48:645–55.[Medline]
13. Ananth CV, Smulian JC, Vintzileos AM. Epidemiology of antepartum fetal testing. Curr Opin Obstet Gynecol 1997; 9:101–6.[Medline]
14. Tricomi V, Kohl SG. Fetal death in utero. Am J Obstet Gynecol 1957;74:1092–7.[Medline]
15. Sheiner E, Hallak M, Shoham-Vardi I, Goldstein D, Mazor M, Katz M. Determining risk factors for intrapartum fetal death. J Reprod Med 2000;45:419–24.[Medline]
16. Harrington K, Cooper D, Lees C, Hecher K, Campbell S. Doppler ultrasound of the uterine arteries: The importance of bilateral notching in the prediction of preeclampsia, placental abruption or delivery of a small-for-gestational-age baby. Ultrasound Obstet Gynecol 1996;7:182–8.[Medline]
17. Lockwood CJ. Heritable coagulopathies in pregnancy. Obstet Gynecol Surv 1999;54:754–65.[Medline]
18. de Vries JIP, Dekker GA, Huijgens PC, Jakobs C, Blomberg BME, van Geijn HP. Hyperhomocysteinaemia and protein S deficiency in complicated pregnancies. Br J Obstet Gynaecol 1997;104:1248–54.[Medline]
19. Kupfermic MJ, Eldor A, Steinman N, Many A, Bar-Am A, Jaffa A, et al. Increased frequency of genetic thrombophilia in women with complications of pregnancy. N Engl J Med 1999;340:9–13.
20. Ananth CV, Savitz DA, Williams MA. Placental abruption and its association with hypertension and prolonged rupture of the membranes: A methodologic review and meta-analysis. Obstet Gynecol 1996;88:309–18.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  M. T. Lydon-Rochelle, V. Cardenas, J. L. Nelson, K. M. Tomashek, B. A. Mueller, and T. R. Easterling Validity of Maternal and Perinatal Risk Factors Reported on Fetal Death Certificates Am J Public Health, November 1, 2005; 95(11): 1948 - 1951. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. C. Fretts, E. B. Elkin, E. R. Myers, and L. J. Heffner Should Older Women Have Antepartum Testing to Prevent Unexplained Stillbirth? Obstet. Gynecol., July 1, 2004; 104(1): 56 - 64. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
