The New England Journal of Medicine
e-mail icon  FREE NEJM E-TOC    HOME   |   SUBSCRIBE   |   CURRENT ISSUE   |   PAST ISSUES   |   COLLECTIONS   |    Advanced Search
Sign in | Get NEJM's E-Mail Table of Contents — Free | Subscribe
 
Original Article
Volume 339:573-577 August 27, 1998 Number 9
NextNext

Reducing the Risk of Multiple Births by Transfer of Two Embryos after in Vitro Fertilization
Allan Templeton, M.D., and Joan K. Morris, Ph.D.

 

This Article
-Abstract
- PDF

Commentary
-Editorial
 by Meldrum, D. R.
-Letters

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation
ABSTRACT

Background In vitro fertilization is associated with a high risk of multiple births, which is a direct consequence of the number of embryos transferred. However, other factors that contribute to the risk are not well defined.

Methods Using the data base established by the Human Fertilisation and Embryology Authority in the United Kingdom, we studied the factors associated with an increased risk of multiple births in 44,236 cycles in 25,240 women. The factors included the woman's age, the cause and duration of infertility, previous attempts at in vitro fertilization, previous live births, number of eggs fertilized, and number of embryos transferred.

Results Older age, tubal infertility, longer duration of infertility, and a higher number of previous attempts at in vitro fertilization were all associated with a significantly decreased chance of a birth and of multiple births. Previous live birth was associated with an increased chance of a birth but not of multiple births. The higher the number of eggs fertilized, the higher the likelihood of a live birth. When more than four eggs were fertilized, there was no increase in the birth rate for women receiving three transferred embryos as compared with those receiving two, but there was a considerable increase in the rate of multiple births when three were transferred (odds ratio, 1.6; 95 percent confidence interval, 1.5 to 1.8).

Conclusions Among women undergoing in vitro fertilization, the chances of a live birth are related to the number of eggs fertilized, presumably because of the greater selection of embryos for transfer. When more than four eggs are fertilized and available for transfer, the woman's chance of a birth is not diminished by transferring only two embryos. Transferring more embryos increases the risk of multiple births.


The high rate of multiple births resulting from in vitro fertilization is a major health issue.1,2,3 The medical, social, and financial consequences are considerable, chiefly because of the excessive morbidity among the survivors of high-order multiple births (triplets or more).4,5,6,7,8 The problem exists because of the large number of embryos transferred in treatment cycles, frequently three or even more. Clinicians and couples are under considerable pressure to maximize pregnancy rates, but there is a view that multiple gestation is an unacceptable consequence of these pressures.5,9 Some have suggested that no more than two embryos need be transferred,10,11 arguing that doing so will not necessarily result in a reduction in pregnancy rates. Others have indicated that transferring only two embryos would be unduly restrictive and would in many instances reduce the chance of success.12,13,14,15

To address this issue, we examined the factors, including the number of embryos transferred, that predispose women undergoing in vitro fertilization to multiple gestation. This study extends a previous analysis of treatment outcomes,16 and like the outcome study it was carried out with the data base established by the Human Fertilisation and Embryology Authority that has recorded all cycles of in vitro fertilization carried out in the United Kingdom since 1991.

Methods

Subjects

We previously studied the factors affecting the outcome of in vitro fertilization treatment, using data from 36,961 cycles in 26,389 women registered in the Human Fertilisation and Embryology Authority data base from 1991 to 1994. Using an extended data base of all 78,325 cycles (in 34,430 women) initiated between August 1991 and April 1995, we analyzed in the current study the factors affecting the likelihood of multiple births.

We excluded from this analysis 2478 cycles with donated eggs in 1689 women, 566 cycles with donated embryos in 398 women, 10,611 transfers of frozen embryos in 7113 women, 4608 unstimulated cycles in 3370 women, and 23,151 cycles in 15,051 women in which one embryo or no embryo was transferred. Because of overlap between the categories, the total numbers of excluded cycles and women were 34,089 and 9190, respectively. Thus, we analyzed 44,236 cycles in 25,240 women. Of these women, 63 percent had one treatment cycle, 20 percent had two, 8 percent had three, and 9 percent had more than three. Their mean (±SD) age was 34±4 years (range, 18 to 52), and the mean duration of infertility was 6±4 years (range, 0 to 20). The indications for in vitro fertilization were tubal disease (50 percent), unexplained infertility (35 percent), endometriosis (10 percent), and other conditions (12 percent); some women had more than one indication. The factors analyzed were the woman's age, the duration and cause of infertility, previous in vitro fertilization treatment, previous live births and whether these were the result of in vitro fertilization, number of eggs fertilized, and number of embryos transferred. The outcome measures were rates of births (live births and perinatal deaths) and multiple births per embryo transfer. Seventy-eight centers contributed both treatment and outcome data for the 44,236 cycles that formed the basis of this analysis. Permission to review the data was granted by the information committee of the Human Fertilization and Embryology Authority.

Statistical Analysis

The relation between age and birth rate was modeled by the method of fractional polynomials,17 on the basis of a logistic-regression analysis that found that a linear relation with age was the most parsimonious. All other factors influencing the probability of a birth or of multiple births were evaluated by logistic-regression analysis with adjustment for age.

Age and duration of infertility were analyzed as continuous variables. Twelve years was taken as the maximal duration; the few women who had been infertile for more than 12 years were assigned a value of 12. The number of eggs fertilized (two only, three or four, or more than four) and previous unsuccessful attempts at in vitro fertilization (no previous attempt, one to three previous attempts, or four or more previous attempts) were analyzed as categorical variables. Both forward and backward selection methods were used to obtain the smallest number of explanatory variables that provided a well-fitting model; the difference in deviance was used to assess whether any additional term was necessary in the model. The confidence intervals for the odds of a birth and of multiple births according to the number of eggs fertilized and the number of embryos transferred were derived from the logistic-regression analysis with use of the method of calculating the floating absolute risk.18

Results

Older age, the presence of tubal infertility, four or more previous attempts at in vitro fertilization, and longer duration of infertility all significantly reduced both the odds of a birth and the odds of multiple births (Table 1). Having had a previous live birth (whether the result of in vitro fertilization or not) increased the odds of a birth but not of multiple births.

View this table:
[in this window]
[in a new window]
 
Table 1. Factors Affecting the Results of in Vitro Fertilization.

 
The effect of the number of embryos transferred was analyzed in relation to the number of eggs fertilized (and thus the number of embryos available for transfer) (Table 1). When only two eggs were fertilized and available for transfer, the odds of a birth, whether single or multiple, were reduced. When more than two eggs were fertilized, the transfer of three embryos did not significantly increase the chance of a birth (over the transfer of two embryos), but it did increase the chances of both twin and triplet births (Table 2).

View this table:
[in this window]
[in a new window]
 
Table 2. Number of Single, Twin, and Triplet Births According to the Number of Eggs Fertilized and of Embryos Transferred, Unadjusted for Age or Other Factors.

 
The odds of a birth were then calculated in relation to the age of the woman, the number of eggs fertilized, and whether two or three embryos were transferred. In all age groups, if more than four eggs were fertilized, the odds of a birth were no greater when three embryos were transferred than when two embryos were transferred (Figure 1). However, the possibility of multiple births did increase (Figure 2). Overall, when age was taken into account, the odds of a birth increased significantly with the number of eggs fertilized, whether two or three embryos were transferred (Figure 3).


View larger version (8K):
[in this window]
[in a new window]
 
Figure 1. Odds of a Birth in Relation to the Woman's Age, Number of Eggs Fertilized, and Number of Embryos Transferred.

The odds of a birth were calculated as compared with those of a 30-year-old woman with more than four eggs fertilized and two embryos transferred (odds for this woman, 1.0).

 

View larger version (7K):
[in this window]
[in a new window]
 
Figure 2. Odds of Multiple Births in Relation to the Woman's Age, Number of Eggs Fertilized, and Number of Embryos Transferred.

The odds of multiple births were calculated as compared with those of a 30-year-old woman with more than four eggs fertilized and two embryos transferred (odds for this woman, 1.0).

 

View larger version (5K):
[in this window]
[in a new window]
 
Figure 3. Odds of a Birth in Relation to the Number of Eggs Fertilized and Number of Embryos Transferred, as Adjusted for Age.

The odds of a birth were calculated as compared with those of a woman with two eggs fertilized. The asterisk denotes P<0.001 for the comparison between two eggs and three or four eggs fertilized, regardless of whether two or three embryos were transferred. The dagger denotes P<0.001 for the comparison between three or four eggs and five or six eggs fertilized, regardless of whether two or three embryos were transferred. The I bars indicate 95 percent confidence intervals.

 
We used these data to construct a table to aid in clinical decision making (Table 3). For this purpose we assumed that each woman had tubal infertility, one to three previous attempts at in vitro fertilization and embryo transfer, and no previous pregnancies. For ages 30, 35, and 40 years, the rates of births (per embryo transfer) and of multiple births (as a percentage of all births) were related to the number of eggs fertilized (and hence available for transfer) and the number of embryos transferred. When three or four eggs were fertilized, the transfer of three embryos, as opposed to two, resulted in a small increase in the birth rate, but at the expense of an increase in the rate of multiple births. However, when more than four eggs were fertilized, the transfer of three embryos did not improve the overall birth rate, but it did result in an increase in the rate of multiple births by 4 to 11 percentage points, depending on age.

View this table:
[in this window]
[in a new window]
 
Table 3. Rates of Births per Embryo Transfer and of Multiple Births as a Percentage of All Births, According to Maternal Age, Number of Eggs Fertilized, and Number of Embryos Transferred.

 
Discussion

In this large study, we found that an increase in the number of embryos transferred invariably resulted in higher rates of multiple births, without necessarily improving the overall success rates of in vitro fertilization. Some studies have suggested that the pregnancy rate after the transfer of two embryos differs little from that when more embryos are transferred.9,10,11 Other studies have suggested that the number of embryos transferred, particularly in older women, can be increased with little risk of multiple births. This assertion has been questioned,19,20 and it is certainly true that in the United Kingdom women over 40 years of age who are undergoing in vitro fertilization and embryo transfer still regularly give birth to twins and triplets.

As previously described in a detailed analysis using the same data base,16 a number of patient characteristics, chiefly the woman's age, diminish the likelihood that a pregnancy will occur after in vitro fertilization. It is now apparent that many of these same characteristics also diminish the chances of multiple gestation. These include treatment for tubal infertility, many previous attempts at in vitro fertilization and embryo transfer, and long duration of infertility. However, in this study, we found that in addition to these patient characteristics, the number of eggs fertilized, and hence the number of embryos available for transfer, was an important factor in determining outcome.

In spite of the wide variation in practices with respect to in vitro fertilization, it appears that in almost all circumstances, and certainly when more than four eggs have been fertilized, the number of embryos available is more important in determining outcome than the number of embryos actually transferred into the uterus. When more than four eggs had been fertilized, the birth rate was similar whether three or two embryos were transferred. Transfer of more embryos, however, increased the risk of multiple births. A 30-year-old woman choosing to have two embryos transferred rather than three reduced her risk of multiple birth from 39 percent to 29 percent and nearly avoided altogether the risk of triplets (Table 3). Similarly, for a 40-year-old woman there was no advantage in transferring three rather than two embryos. All that was achieved was an increase in the rate of multiple births from 23 percent to 27 percent.

These results come from centers with wide variations in clinical practice. The low success rate when only two eggs were fertilized reflects the lack of choice among embryos for transfer, whereas the increased choice when more than four eggs were fertilized is reflected in higher pregnancy rates. Thus, the ability to select embryos remains an important determinant of the outcome of in vitro fertilization, despite the limitations of grading the quality of embryos according to morphologic criteria, as is done in current clinical practice. When three or even more embryos selected from a larger number are transferred, there will inevitably be a higher rate of multiple births.

The data presented here are also consistent with the suggestion that uterine receptivity is likely to be an all-or-nothing effect.21 When the uterus is receptive and the embryos have good implantation potential, as seems more likely when they are chosen from a large number, transfer of three will result in a higher chance of twins or triplets than transfer of two. On the other hand, if the uterus is not receptive, a pregnancy will not occur, no matter how many embryos are transferred. Centers in which the embryos have a higher potential for implantation may be more likely to reduce the number of embryos transferred in order to minimize the rate of multiple births, whereas centers with lower pregnancy rates may continue to transfer three embryos in order to increase their pregnancy rates. However, this pattern does not explain the significantly higher rates of multiple births when three embryos were transferred, nor was there any evidence that this effect was a contributing factor when the transfer policies of the individual centers were reviewed.

The implications of these results for clinical practice are clear. When more than four embryos are fertilized and available, transfer of only two embryos will not diminish the woman's chance of becoming pregnant, but it will reduce her chance of multiple births. This is true for women of all ages. When only three or four embryos are fertilized and the woman is about 40 years old, there may be a slight increase in the pregnancy rate when three embryos are transferred, but the risk of multiple births will also increase. When only two embryos are available for transfer, these issues do not arise, but the couple should be prepared to accept a much lower pregnancy rate.

We are indebted to the staff of the Human Fertilisation and Embryology Authority for providing the data for this analysis, in particular to Mr. Bill Parslow, Miss Joanne Thompson, and Mr. Richard Baranowski; to Dr. Anne McLaren for her review of the manuscript and helpful comments; and to the individual clinics registered with the Human Fertilisation and Embryology Authority for data collection.


Source Information

From the Human Fertilisation and Embryology Authority, London (A.T.); and the Department of Environmental and Preventive Medicine, Wolfson Institute of Preventive Medicine, London (J.K.M.).

Address reprint requests to Professor Templeton at the Department of Obstetrics and Gynaecology, Aberdeen Maternity Hospital, Cornhill Rd., Aberdeen AB25 2ZD, Scotland.

References

  1. Society for Assisted Reproductive Technology, American Society for Reproduction Medicine. Assisted reproductive technology in the United States and Canada: 1993 results generated from the American Society for Reproductive Medicine/Society for Assisted Reproductive Technology Registry. Fertil Steril 1995;64:13-21. [Medline]
  2. Pregnancies and births resulting from in vitro fertilization: French national registry, analysis of data 1986 to 1990: FIVNAT (French In Vitro National). Fertil Steril 1995;64:746-756. [Medline]
  3. Sixth annual report: 1997. London: Human Fertilisation and Embryology Authority, 1997.
  4. Bronson R. How should the number of embryos transferred to the uterus following in-vitro fertilization be determined to avoid the risk of multiple gestation? Hum Reprod 1997;12:1605-1607. [Free Full Text]
  5. Faber K. IVF in the US: multiple gestation, economic competition, and the necessity of excess. Hum Reprod 1997;12:1614-1616. [Medline]
  6. Callahan TL, Hall JE, Ettner SL, Christiansen CL, Greene MF, Crowley WF Jr. The economic impact of multiple-gestation pregnancies and the contribution of assisted-reproduction techniques to their incidence. N Engl J Med 1994;331:244-249. [Free Full Text]
  7. Baldwin VJ. Pathology of multiple pregnancy. New York: Springer-Verlag, 1994.
  8. Jones HW Jr. Twins or more. Fertil Steril 1995;63:701-702. [Medline]
  9. Roest J, van Heusden AM, Verhoeff A, Mous HV, Zeilmaker GH. A triplet pregnancy after in vitro fertilization is a procedure-related complication that should be prevented by replacement of two embryos only. Fertil Steril 1997;67:290-295. [CrossRef][Medline]
  10. Staessen C, Janssenswillen C, Van den Abbeel E, Devroey P, Van Steirteghem AC. Avoidance of triplet pregnancies by elective transfer of two good quality embryos. Hum Reprod 1993;8:1650-1653. [Free Full Text]
  11. Tasdemir M, Tasdemir I, Kodama H, Fukuda J, Tanaka T. Two instead of three embryo transfer in in-vitro fertilization. Hum Reprod 1995;10:2155-2158. [Free Full Text]
  12. Craft I, al-Shawaf T. Limiting the number of oocytes and embryos transferred in GIFT and IVF. BMJ 1990;303:185-185. 
  13. Cohen J. The efficiency and efficacy of IVF and GIFT. Hum Reprod 1991;6:613-618. [Free Full Text]
  14. Widra EA, Gindoff PR, Smotrich DB, Stillman RJ. Achieving multiple-order embryo transfer identifies women over 40 years of age with improved in vitro fertilization outcome. Fertil Steril 1996;65:103-108. [Medline]
  15. Azem F, Yaron Y, Amit A, et al. Transfer of six or more embryos improves success rates in patients with repeated in vitro fertilization failures. Fertil Steril 1995;63:1043-1046. [Medline]
  16. Templeton A, Morris JK, Parslow W. Factors that affect outcome of in-vitro fertilisation treatment. Lancet 1996;348:1402-1406. [CrossRef][Medline]
  17. Royston P, Altman DG. Regression using fractional polynomials of continuous covariates: parsimonious parametric modeling. Appl Stat 1994;43:429-67.
  18. Easton DF, Peto J, Babiker AG. Floating absolute risk: an alternative to relative risk in survival and case-control analysis avoiding an arbitrary reference group. Stat Med 1991;10:1025-1035. [Medline]
  19. Walters DE. The statistical implication of the `number of replacements' in embryo transfer. Hum Reprod 1996;11:10-12. [Free Full Text]
  20. Senoz S, Ben-Chetrit A, Casper RF. An IVF fallacy: multiple pregnancy risk is lower for older women. J Assist Reprod Genet 1997;14:192-198. [Medline]
  21. McLaren A. The control of implantation. In: Thompson W, Joyce DN, Newton JR, eds. In vitro fertilization and donor insemination: proceedings of the Twelfth Study Group of the Royal College of Obstetricians and Gynaecologists. London: Royal College of Obstetricians and Gynaecologists, 1985:13-22.

 

This Article
-Abstract
- PDF

Commentary
-Editorial
 by Meldrum, D. R.
-Letters

Tools and Services
-Add to Personal Archive
-Add to Citation Manager
-Notify a Friend
-E-mail When Cited

More Information
-PubMed Citation

Related Letters:

Estimating the Risk of Cerebral Palsy after Assisted Conceptions
Blickstein I., Weissman A.
Extract | Full Text  
N Engl J Med 1999; 341:1313-1314, Oct 21, 1999. Correspondence

This article has been cited by other articles:



HOME  |  SUBSCRIBE  |  SEARCH  |  CURRENT ISSUE  |  PAST ISSUES  |  COLLECTIONS  |  PRIVACY  |  HELP  |  beta.nejm.org

Comments and questions? Please contact us.

The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved.