MacKay, Andrea P. MSPH1; Berg, Cynthia J. MD, MPH2; King, Jeffrey C. MD3; Duran, Catherine1; Chang, Jeani2
The number and rate of multiple births have increased dramatically during the past 2 decades, in part due to trends in older child bearing and increased use of fertility-enhancing therapies.1,2 Between 1979 and 2000, the twin birth rate increased 55%, and the rate for triplet and other higher-order births increased 388%. Growth in multifetal birth rates was most marked among women aged 30 years and older. Multifetal pregnancies are of clinical and public health concern because of the elevated risks for the mother and child. Infants born in multiple deliveries have a higher incidence of preterm birth and low birthweight and are at a greater risk of infant death.3–5 Similarly, multifetal pregnancies are associated with an increased risk of morbidity for the mother, including increased rates of preeclampsia, cardiac morbidity, amniotic fluid or thromboembolism, the need for obstetric intervention, postpartum hemorrhage, hysterectomy, blood transfusion, prolonged hospital stay, and pregnancy-related death.5,6
We used data from the Centers for Disease Control and Prevention’s (CDC) Pregnancy Mortality Surveillance System to conduct an analysis of pregnancy-related mortality associated with multifetal pregnancies resulting in a live birth or fetal death. We reviewed all reported pregnancy-related deaths in the United States for 1979–2000 to understand trends in pregnancy-related mortality associated with multifetal pregnancies, to identify risk factors for these deaths, and to calculate plurality-specific pregnancy mortality ratios.
MATERIALS AND METHODS
The Centers for Disease Control and Prevention’s Division of Reproductive Health, in collaboration with the American College of Obstetricians and Gynecologists and state health departments, conducts ongoing surveillance of pregnancy-related deaths. Offices of vital registration in the 50 states, New York City, and the District of Columbia provided CDC with copies of death certificates for all identified pregnancy-related deaths. The Centers for Disease Control and Prevention also requested the matching birth or fetal death certificates for all deaths that followed a live birth or stillborn fetus. Beginning with deaths occurring in 1991, states were asked to send certificates of all deaths that occurred during or within 1 year of the end of pregnancy, regardless of the cause of death. This analysis includes data from 1979–2000, the most recent years for which completed data are available.
Deaths were considered pregnancy-related if they occurred during pregnancy or within 1 year of the termination of pregnancy and resulted from complications of pregnancy itself, a chain of events initiated by pregnancy, or aggravation of an unrelated event by the physiologic effects of pregnancy.7 Information on all death certificates provided by the States was reviewed by clinically experienced epidemiologists to determine whether the death was pregnancy-related and coded for cause of death, contributing conditions of death, and outcome of the pregnancy. Deaths were classified using the system designed in collaboration with the American College of Obstetricians and Gynecologists/CDC Maternal Mortality Study Group and initiated in 1988.7,8 Surveillance was conducted retrospectively for 1979 to 1987, and prospectively each year after. Increases in overall pregnancy-related mortality since 1987 have been attributed to improved surveillance and reporting guidelines.8–10
Based on a review of all clinical information available for each death, pregnancy-related deaths were classified into 1 of 8 cause-of-death groups: hemorrhage, infection, embolism, hypertensive disorders of pregnancy, anesthesia complications, cardiomyopathy, cerebrovascular accidents, or other medical conditions (excludes external causes such as suicide).
For this analysis, we included all identified pregnancy-related deaths with a live birth or fetal death outcome. From information available, pregnancies of the women who died were identified as either singleton or multifetal. In the absence of this information, the pregnancy was considered a singleton. Approximately 7% of deaths considered singleton did not have matched live birth or fetal death certificates, which might have provided additional information on plurality.
To assess the risk of pregnancy-related death associated with multifetal pregnancy, we used a pregnancy-based mortality ratio, defined as the ratio of the number of pregnancy-related deaths to the number of pregnancies with a live birth. Plurality-specific mortality ratios were calculated separately for singleton and multifetal pregnancies. For multifetal pregnancies the pregnancy-based mortality ratio was defined as the number of pregnancy-related deaths among women with multifetal (twin and higher-order) pregnancies per 100,000 multifetal (twin and higher-order) pregnancies with a live birth. For singleton pregnancies the pregnancy-based mortality ratio was defined as the number of pregnancy-related deaths among women with singleton pregnancies per 100,000 singleton pregnancies with a live birth.
Information on singleton and twin and higher-order live births in the United States was obtained from CDC’s National Center for Health Statistics.11 For 1979–2000, no national data system reported the number of pregnancies with a live birth outcome. Therefore, we estimated the number of pregnancies with a live birth for each year using the number of live births. To estimate the number of twin pregnancies, we divided the number of twin live births by 2, and for triplet and higher-order pregnancies we divided triplet and higher-order live births by 3. The estimated numbers of twin and triplet and higher-order pregnancies were combined to provide the number of multifetal pregnancies with a live birth. Although this method is not fully weighted to account for quadruplet and higher births, those births are very rare and would have little effect on the total. Although we recognize that some multifetal pregnancies result in an outcome that includes both (1 or more) live births and stillbirths, we believe this estimate closely approximates the number of multiple pregnancies with live births.
Because of the relatively small number of deaths among women with multifetal pregnancies, we calculated 3-year (average annual) rolling mortality ratios. The relative risk (RR) of pregnancy-related death was calculated for women with multifetal pregnancies compared with women with singleton pregnancies. Ninety-five percent confidence intervals and P values were calculated using Epi Info.12
Plurality-specific pregnancy mortality was analyzed by race, age group, education, marital status, and cause of death. Race-specific mortality was not analyzed for women classified as other than white or African American because of small numbers of pregnancy-related deaths. Numbers of deaths were also too small to analyze pregnancy-related mortality for Hispanic women. Age groups were defined as less than 20 years, 20–29 years, 30–34 years, and 35 years and older. Because of small numbers, in the analysis of results by age and race, only 2 age groups were used: younger than 30 years and 30 years and older. The analysis of education was limited to women aged 20 years or older, an age by which most women would have had the opportunity to graduate from high school. The analysis of education was further restricted to those states which reported educational level on the birth certificate (42 states and the District of Columbia in 1979, 47 states and DC in 1980–87, 46 states and the District of Columbia in 1988, 48 states and the District of Columbia in 1989–91, and all states and the District of Columbia in 1992–2000). For this analysis, women who were divorced or widowed were classified as unmarried. Ratios calculated for multifetal pregnancy-related deaths are based on relatively small numbers, and the point estimates for the stratified analysis should be interpreted with caution.
We grouped deaths into 2 periods (1979–89 and 1990–2000), by plurality, to evaluate secular trends. In particular, we wanted to identify any potential increased risk among women who were characteristic of those most likely to use fertility-enhancing therapies. This study was exempt from Institutional Review Board review because it did not involve research using living human subjects.
For the 22-year study period, we identified 4,992 pregnancy-related deaths to women with a live birth or fetal death; 4.2% of these (209 deaths) were among women with multifetal pregnancies. Overall, the plurality-specific pregnancy mortality ratio among women with multifetal pregnancies was 20.8 deaths per 100,000 multifetal pregnancies compared with a ratio among women with singleton pregnancies of 5.8 deaths per 100,000 singleton pregnancies. Reported mortality ratios for women with singleton pregnancies increased fairly steadily over the study period, from approximately 3 deaths per 100,000 pregnancies to 9 per 100,000 (Fig. 1). The trend in mortality ratios for women with multifetal pregnancies fluctuated and also increased over the study period and ranged from 11 to 36 deaths per 100,000 pregnancies. Ascertainment of all pregnancy-related deaths improved over this period and likely explains much of this increase. The RR of pregnancy-related death associated with multifetal pregnancy compared with singleton pregnancy during the study period was 3.6, ranging from 2.1 to 4.4; this RR was similar for the 2 decades of the study period (3.7 compared with 3.4, respectively).
The most frequent causes of pregnancy-related death among singleton and multifetal pregnancies were similar: embolism, hemorrhage, hypertensive disorders of pregnancy, and infection. However, women with multifetal pregnancies were 3 to 4 times as likely to die from these causes as women with singleton pregnancies (Table 1). The RR of pregnancy-related death among women with multifetal pregnancies compared with those with singleton pregnancies was 4.2 for embolism, 3.9 for hemorrhage, and 3.1 for hypertensive disorders. The RR was essentially the same for most causes of death during the 2 periods of the study 1979–89 and 1990–2000 (data not shown). During the 22-year study period, fewer than 5 deaths among women with multifetal pregnancies were attributed to complications of anesthesia and cerebrovascular accidents.
The risk of death from pregnancy was higher for older women than for younger women, regardless of the plurality of the pregnancy (Table 2). However, for women aged 30 years and older, as maternal age increased, the RR of pregnancy death associated with multifetal pregnancy compared with singleton pregnancy decreased. The RR associated with multifetal pregnancy was similar for both African-American and white women. For both singleton and multifetal pregnancy-related deaths, African-American women were about 3 times as likely to die as white women. The RR for each age, race, education, and marital status category for women with multifetal pregnancies compared with women with singleton pregnancies were similar during the 2 study periods (data not shown).
The risk of pregnancy-related death was greater among women with multifetal pregnancies compared with that of women with singleton pregnancies at every level of educational attainment (4.1, 3.6, and 3.5, respectively) (Table 2). It was also greater for both married and unmarried women with multifetal pregnancies than for their counterparts with singleton pregnancies (Table 2).
Women with multifetal pregnancies have a significantly higher risk of pregnancy-related death than their counterparts with singleton pregnancies; this holds true for all women regardless of age, race, level of education, and marital status. The 3- to 4-fold disparity between African-American and white women in overall pregnancy-related mortality was also observed among women with twin and higher-order pregnancies.8,10 Improved ascertainment of all pregnancy-related deaths contributed to increases in pregnancy-related mortality ratios for both singleton and multifetal pregnancies between the 2 time of the study.9 The ratio most commonly used to assess and report pregnancy-related mortality and maternal mortality is deaths per 100,000 live births7–10; however, in this analysis the plurality-specific pregnancy-based mortality ratio was used to provide a more accurate assessment of the risk of pregnancy-related death for women with multifetal pregnancies compared with women with singleton pregnancies. This unique ratio weights the risk per pregnancy rather than per live birth. Although the absolute risks for pregnancy-related death among women with multifetal and singleton pregnancies may seem low, most pregnancy-related deaths are preventable.9
The leading causes of pregnancy-related death among women with multifetal pregnancy were similar to those of women with singleton pregnancy: embolism, hemorrhage, hypertensive disorders of pregnancy, and infection. The normal physiologic changes associated with pregnancy, such as increased plasma volume, increased cardiac output, and changes in coagulation factors, are further heightened in multifetal pregnancies,6 and significantly higher rates of preeclampsia, embolism, and hemorrhage in multifetal pregnancies have been reported.5,6 Increased medical morbidity, obstetric complications, and surgical interventions have been associated with multifetal pregnancies. Women with multifetal pregnancies are nearly 6 times more likely to be hospitalized with complications, including preeclampsia, pyelonephritis, and postpartum hemorrhage.4 Risk factors associated with thromboembolism (maternal age of 35 years or older, body mass index of 25 or higher, and cesarean delivery) are more common in multifetal pregnancies.13 In addition, women with multifetal pregnancies are often placed on bed rest, and the enlarged uterus mechanically obstructs venous return, contributing to lower extremity stasis. The incidence of preeclampsia is 2.6 times higher in twin pregnancies than in singleton pregnancies.14 Further, when multifetal pregnancy is complicated by preeclampsia, it is significantly more likely to occur earlier and be more severe.14,15 These myriad exaggerated physiologic effects are reflected in the increased risk of mortality from all causes for women with multifetal pregnancies.
The limitations of this analysis should be considered. Despite improved ascertainment methods initiated over the study period, pregnancy-related mortality ratios are underestimated; up to one half of pregnancy-related deaths are not identified through routine surveillance methods.16,17 Further, matched live birth or fetal death certificates were not available for 7% of pregnancy-related deaths. Because other specific information indicative of plurality was not evident, these were classified as singleton. Misclassification of multifetal pregnancies as singleton pregnancies would underestimate the multifetal pregnancy mortality ratio and overestimate the singleton pregnancy mortality ratio. The relatively small numbers of multifetal deaths did not allow detailed stratification by more than 1 variable at a time. In addition, information for multifetal pregnancy-related deaths did not always indicate the specific plurality of multiples. Therefore, we could not determine the RR of pregnancy-related death between twin pregnancies and higher-order multifetal pregnancies. However, the risk of maternal morbidity and obstetric complications in triplet and higher-order pregnancies is greater than in twin pregnancies.4,18,19
The use of assisted reproductive technology (ART) procedures increased dramatically during the 1990s.2 Pregnancies associated with ART or ovulation-inducing drugs are more likely to result in multiple births than spontaneously conceived pregnancies. In 2000, 42% of triplet and higher-order births were a result of ART, 40% were a result of ovulation-inducing drugs, and 18% were the result of spontaneous conception.2,20 Compared with women with singleton pregnancies, women with twin and triplet pregnancies are more likely to be older, non-Hispanic white, married, and better-educated.21 Concern has been expressed that multifetal pregnancies conceived through fertility-enhancing therapies may carry higher risks for the mother than spontaneous multifetal pregnancies. Information on the use of assisted reproductive technology among the women who died was not available; therefore, we were unable to determine directly whether assisted multifetal pregnancies were at any higher risk of pregnancy-related death than naturally occurring multifetal pregnancies. However, when we stratified by age and race, we did not find an increase in the RR of pregnancy-related death for white women aged 30 years and older with multifetal pregnancies between the 1980s and the 1990s (ie, the group expected to have the greatest use of ART) (data not shown). Although the specific contribution to the risk of pregnancy-related death cannot be assessed for fertility-enhancing therapies, the marked increase in multifetal pregnancies due to these therapies contributes to the cohort of women at an elevated risk of pregnancy-related death.
The American Society for Reproductive Medicine has developed and promulgated recommendations related to the optimal number of embryo transfers based upon maternal age and the presumed cause of the infertility.22 These guidelines, along with a transition to later embryo transfer, have dramatically reduced the number of higher-order multifetal gestations in more recent years. The twin rate for ART patients increased between 1997 and 2000, reaching 444.7 per 1,000 live births in 2000, whereas the triplet and higher-order birth rate for ART patients declined substantially from 134.3 to 98.7 per 1,000 live births from 1997 to 2000.20 Now, most higher-order multifetal gestations result from superovulation or ovulation induction rather than in vitro fertilization.23 In recent years, there has been an increasing use of single embryo transfer to further control the resultant twin and higher-order multifetal gestations.24
This study provides an analysis of the pregnancy-related mortality risk associated with multifetal pregnancies based on national data. These findings identify the significantly increased risk of death for women carrying more than a single fetus irrespective of maternal age, race, marital status, or level of education. Further research into the specific factors associated with the substantially increased pregnancy-related mortality ratio for women with multifetal pregnancies is clearly indicated.
1. Martin JA, Park MM. Trends in twin and triplet births:1980-97. Natl Vital Stat Rep 1999;47:1–16.
2. Wright VC, Schieve LA, Reynolds MA, Jeng G. Assisted reproductive technology surveillance—United States, 2000. MMWR Surveill Summ 2003;52:1–16.
3. Kiely JL. What is the population-based risk of preterm birth among twins and other multiples?. Clin Obstet Gynecol 1998;41:3–11.
4. Luke B, Keith LG. The contribution of singletons, twins and triplets to low birth weight, infant mortality and handicap in the United States. J Reprod Med 1992 Aug;37:661–6.
5. Multiple gestation: complicated twin, triplet, and high-order multifetal pregnancy. ACOG Practice Bulletin No. 56. American College of Obstetricians and Gynecologists; Society for Maternal-Fetal Medicine. Obstet Gynecol 2004;104:869–83.
6. Walker MC, Murphy KE, Pan S, Yang Q, Wen SW. Adverse maternal outcomes in multifetal pregnancies. BJOG 2004;111:1294–6.
7. Ellerbrock TV, Atrash HK, Hogue CJ, Smith JC. Pregnancy mortality surveillance: a new initiative. Contemp Ob Gyn 1988;31:23–34.
8. Berg CJ, Chang J, Callaghan WM, Whitehead SJ. Pregnancy-related mortality in the United States, 1991–1997. Obstet Gynecol 2003;101:289–96.
9. Berg C, Danel I, Atrash H, Zane S, Bartlett L. Strategies to reduce pregnancy-related deaths: from identification and review to action. Atlanta (GA): Centers for Disease Control and Prevention; 2001. p. 13–7.
10. Koonin LM, MacKay AP, Berg CJ, Atrash HK, Smith JC. Pregnancy-related mortality surveillance—United States, 1987–1990. MMWR CDC Surveill Summ 1997;46:17–36.
11. National Center for Health Statistics. Natality public-use tapes 1979–2000. Hyattsville (MD): National Center for Health Statistics. Annually 1981–2002.
12. Epidemiology Program Office, Division of Public Health Surveillance and Informatics, Centers for Disease Control and Prevention. Epi Info, Version 3.3.2. Atlanta (GA): 2005.
13. Simpson EL, Lawrenson RA, Nightingale AL, Farmer RD. Venous thromboembolism in pregnancy and the puerperium: incidence and additional risk factors from a London perinatal database. BJOG 2001;108:56–60.
14. Sibai BM, Hauth J, Caritis S, Lindheimer MD, MacPherson C, Klebanoff M, et al. Hypertensive disorders in twin versus singleton gestations. National Institute of Child Health and Human Development Network of Maternal-Fetal Medicine Units. Am J Obstet Gynecol 2000;182:938–42.
15. Mastrobattista JM, Skupski DW, Monga M, Blanco JD, August P. The rate of severe preeclampsia is increased in triplet as compared to twin gestations. Am J Perinatol 1997 May;14:263–5.
16. Centers for Disease Control and Prevention (CDC). Pregnancy-related mortality–Georgia, 1990-1992. MMWR Morb Mortal Wkly Rep 1995;44:93–6.
17. Colorado Maternal Mortality Review Committee Brief. Denver (CO): Colorado Department of Public Health and Environment; Aug 2000, No. 1.
18. Wen SW, Demissie K, Yang Q, Walker MC. Maternal morbidity and obstetric complications in triplet pregnancies and quadruplet and higher-order multiple pregnancies. Am J Obstet Gynecol 2004;191:254–8.
19. Aliyu MH, Salihu HM, Blankson ML, Alexander GR, Keith L. Risks in triplet pregnancy: advanced maternal age, premature rupture of membranes and risk estimates of mortality. J Reprod Med 2004;49:721–6.
20. Reynolds MA, Schieve LA, Martin JA, Jeng G, Macaluso M. Trends in multiple births conceived using assisted reproductive technology, United States, 1997–2000. Pediatrics 2003 May;111:1159–62.
21. Zhang J, Meikle S, Grainger DA, Trumble A. Multifetal pregnancy in older women and perinatal outcomes. Fertil Steril 2002;78:562–8.
22. Practice Committee of the Society for Assisted Reproductive Technology and the American Society for Reproductive Medicine. Guidelines on the number of embryos transferred. Fertil Steril 2004;82 suppl:1–2.
23. The Practice Committee of the American Society for Reproductive Medicine. Multiple pregnancy associated with infertility therapy. Fertil Steril 2004;82 suppl:153–7.
24. Thurin A, Hausken J, Hillensjo T, Jablonowska B, Pinborg A, Strandell A, et al. Elective single-embryo transfer versus double-embryo transfer in in vitro fertilization. N Engl J Med 2004;351:2392–402.