| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From Tornblad Institute, University of Lund, Lund, Sweden; Centre for Epidemiology, National Board of Health and Welfare, Stockholm, Sweden; and Sophiahemmet, Stockholm, Sweden.
Address reprint requests to: B. Källén, MD, PhD, University of Lund, Tornblad Institute, Biskopsgatan 7, S-223, 62 Lund, Sweden; E-mail: embryol{at}embryol.lu.se.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: We identified 4029 women who delivered between 1995–1999 after ovarian stimulation alone and compared them with 438,582 women who neither had ovarian stimulation nor in vitro fertilization. We controlled for the confounding effect of year of birth, maternal age, parity, and length of subfertility before the pregnancy.
RESULTS: The twinning rate was 5.9% in the study group and 1.2% in the control group. The triplet rate was 0.5% in the study group and 0.02% in the control group. A nearly doubling of the rate of monozygotic twinning was indicated in the study group compared with the control group. There was an excess of singleton preterm births and low birth weight infants in the study group, but this was mainly explainable by confounding of maternal age, parity, and subfertility. The rates of congenital malformations and perinatal deaths were increased, also mainly explainable by maternal characteristics. No increase in specific types of congenital malformations was seen.
CONCLUSION: As the deviations in neonatal outcome after ovarian stimulation alone were reduced or disappeared when the confounding of maternal age, parity, and subfertility was taken into consideration, there is probably little direct effect of the stimulation procedure as such.
A long debate has been held on the possible teratogenic effect of ovarian stimulation, notably the use of clomiphene citrate. This question has received a renewed interest in relation to in vitro fertilization (IVF, including intracytoplasmic sperm injection) pregnancies where ovarian stimulation is often used. In the present study, we investigated the neonatal outcome of infants born in pregnancies resulting from ovarian stimulation not including IVF and compared it with that seen among infants born after spontaneous pregnancies. Our object was to see whether the risk of multiple births, of preterm or low birth weight in singleton infants, of infants with congenital malformations, or of infants who died was increased after ovarian stimulation.
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We identified all women who were delivered in 1995–1999 with ovarian stimulation but without IVF. Presence of IVF was further checked by linkage with a national register of women who have undergone IVF.2–4 That register is built from direct reports from all clinics performing IVF in Sweden and is thought to be nearly complete for births during the period 1982–1997.
We compared the study group (women who had ovarian stimulation) with all women giving birth with pregnancies without ovarian stimulation or IVF. Year of delivery, maternal age, parity, and subfertility were regarded as confounding factors.
Data on maternal age, parity, subfertility, pregnancy duration, and birth weight were retrieved from the Medical Birth Register. Data on congenital malformations were obtained both from the Medical Birth Registry and from the Swedish Registry of Congenital Malformations.5 Information on child death was obtained from the Medical Birth Registry, which contains information on stillbirths (according to the Swedish definition of more than 28 completed gestational weeks) and date of death of a live-born infant (obtained by linkage with the official register of death in Sweden).
We first investigated how maternal age, parity, and subfertility affected the probability to have had an ovarian stimulation therapy. We then compared neonatal outcome in the ovarian stimulation group with the control group, with and without consideration to subfertility as a confounder.
To estimate the proportion of monozygotic and dizygotic twins, the Weinberg law was applied. This law states that the number of like-sexed dizygotic twins is the same as the number of unlike-sexed dizygotic twins. By subtracting the number of unlike-sexed twins from the number of like-sexed twins, the number of monozygotic twins can be estimated. Zygosity of like-sexed individual twin pairs was not known.
Comparisons were made between infants born after ovarian stimulation and infants born in pregnancies without ovarian stimulation or IVF during the period July 1, 1995, to 1999. Odds ratios (ORs) were determined using Mantel-Haenszel analysis after various stratifications for confounders. Approximate 95% confidence intervals (CIs) were estimated with a test-based method.6
To compare observed and expected numbers of infants with specific malformations, exact Poisson distributions were used (SABER Software, Centers for Disease Control and Prevention, Atlanta, GA).
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The women undergoing ovarian stimulation were older and more often of parity 1 than the control women (Table 1
). When the length of involuntary childlessness as a measurement of subfertility was compared between subfertile women who got ovarian stimulation and who did not, a maximum OR for having an ovarian stimulation was seen after 3 years of childlessness.
|
We compared the observed and expected number of like-sexed and unlike-sexed twin pairs among 235 twin pairs with known sex, taking year of birth, maternal age and parity, and years of involuntary childlessness into consideration. There were 133 like-sexed pairs and 102 unlike-sexed pairs. The expected numbers were 31.8 and 14.0, respectively. Applying Weinberg’s law, it could be estimated that 31 of the like-sexed pairs were monozygotic and the expected number was 17.8, giving a risk ratio of 1.74.
Table 2 summarizes the neonatal outcome. Preterm birth and low birth weight in singletons occurred in excess with ORs of 1.39 and 1.69, respectively, when no adjustment for subfertility was made. After stratification also for number of years of involuntary childlessness, the ORs decreased. This was still more apparent when only women with stated length of childlessness were analyzed. However, significantly increased risks remained for both preterm birth and low birth weight also after adjustment for the length of involuntary childlessness.
|
). Twenty-three of the 41 perinatally dead infants born after ovarian stimulation were stillborn, 11 of the stillbirths were preterm, and none had a congenital malformation. Among the 18 live-born infants who died before the age of 7 days, 13 were preterm, ten of them born before 30 weeks’ gestation. Four of these infants belonged to two twin pairs, born after 23 completed weeks.
Three of the infants born after ovarian stimulation who died before 7 days of age had congenital malformations (marked in Tables 3 and 4). Another seven infants died during the first month of life and another two during the first year of life. Two of the children died from sudden infant death syndrome.
|
|
). There were 187 malformed infants identified from the Medical Birth Registry and a further ten from the Registry of Congenital Malformations.
Table 3 specifies all identified major congenital malformations, and Table 4 shows mild or variable anomalies. Eight infants had more than one malformation (Table 5).
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In a recent analysis of infants born in Sweden after IVF, it was demonstrated that the crude increase in risk for congenital malformations was explained by confounding of maternal age, parity, and subfertility.15 However, when specific types of congenital malformations were analyzed, excess risks among IVF infants were seen for neural tube defects; esophageal, gut, or anal atresia; hypospadias (only after intracytoplasmatic sperm injection); and possibly omphalocele. In IVF treatments, ovarian stimulation is usually combined with the in vitro procedures, which may increase the risk of certain malformations.
The cohort of births after ovarian stimulation without IVF, which was analyzed in this paper, was identified in the nationwide Swedish Medical Birth Register. An IVF treatment is not always marked in the records, but data were supplemented from a specific register of IVF pregnancies up to and including 1997. A few women who have had IVF and gave birth in 1998 or 1999 may have been included in the study, but most likely they are few. Exposure information was prospective related to neonatal outcome.
The most clear-cut result was the high rate of multiple births after ovarian stimulation. The twin delivery rate was 5.9% compared with 1.2% among spontaneous pregnancies, and the risk increase for triplets was still higher. In a previous study from Sweden,9 seven twin deliveries were found among 141 deliveries after clomiphene citrate therapy (5%). On the other hand, a 25-fold increase in twinning rate both after ovarian stimulation and after IVF was recently reported.16
It has previously been shown that the total twinning rate in Sweden has increased markedly since the 1980s, even after correction for the changing maternal age distribution.3,4 About half of the increase was due to IVF pregnancies (where the twinning rate is 25% in 1995–1997), and the remaining increase has been thought to be due to ovarian stimulation alone. As the twinning rate after ovarian stimulation was 5.9%, that would mean that there should be about four times more pregnancies with ovarian stimulation alone than there were IVF pregnancies. The rate of IVF pregnancies in the population in 1995–1997 was about 1.3%,4 which would mean that the rate of deliveries after ovarian stimulation alone in the population is about 5% instead of the 1.3% found in the present study, an under-reporting, which seems rather unlikely. Another possibility is that other factors have contributed to the increase in twinning rate. One possible factor, which may increase the twinning rate was recently demonstrated: the increasing use of folic acid periconceptionally.15
The increased twinning rate after ovarian stimulation alone was mainly due to an increase in dizygotic twinning, but also monozygotic twinning appears to be increased in rate. This phenomenon has been demonstrated in IVF pregnancies.2,17,18 It was also described after ovarian stimulation in the absence of IVF and was suggested to be due to ovarian stimulation and not to IVF procedures.18
Comparisons of pregnancy outcome were made between the ovarian stimulation cohort and women without ovarian stimulation or IVF pregnancy. After stratification for year of birth, maternal age, and parity, an increased OR was found for all variables studied: preterm birth, low birth weight, perinatal death, and congenital malformations. When further adjustment was made for subfertility, evaluated as the known length of involuntary childlessness, the ORs decreased and for congenital malformations, the increased risk was not any longer statistically significant. The adjustment for subfertility may not completely remove its confounding effect: one will compare women who had ovarian stimulation with women who, after a similarly long period of infertility, became spontaneously pregnant. On the other hand, the ORs will be biased towards 1.0 because of the presence of unidentified women who had IVF or ovarian stimulation alone in the reference population, of most significance in the subgroups with subfertility.
The nature of a possible link between subfertility and neonatal problems is in most instances unknown.
The apparent effect of ovarian stimulation on congenital malformation rate is probably explainable by a confounding effect of maternal subfertility. In deliveries occurring after IVF, we found an OR for any congenital malformation of 1.47,2 which is higher than that found after ovarian stimulation in the present investigation. In both groups, the ORs became close to 1 when adjustment for subfertility was made, and it seems likely that the increase in risk is caused by factors related to subfertility, which may differ between women who had IVF and women who only had ovarian stimulation.
There is no clear-cut increase in any specific major malformation after ovarian stimulation, and no infant had a neural tube defect. It is possible, however, that women who have had ovarian stimulation alone underwent more intense prenatal diagnostic efforts than other women, which could result in a reduction of the frequency of severe malformations among their infants.
In conclusion, we found that pregnancies resulting from ovarian stimulation alone have a less favorable outcome than spontaneous deliveries with respect to preterm birth, low birth weight, presence of congenital malformations, and perinatal death. We found no specificity with respect to the type of congenital malformation. The mechanism causing this suboptimum neonatal outcome is probably related to the underlying subfertility.
|
Footnotes |
|---|
Received December 7, 2001. Received in revised form March 5, 2002. Accepted April 4, 2002.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Ericson A, Källén B. Congenital malformations in infants born after in vitro fertilization. A population based study. Hum Reprod 2000;16:504–9.
3. Bergh T, Ericson A, Hillensjö T, Nygren K-G, Wennerholm U-B. Delivery and children born after in-vitro fertilisation in Sweden 1982–1995: A retrospective cohort study. Lancet 1999;354:1579–85.[Medline]
4. EpC Report. Barn födda i Sverige efter provrörsbefruktning, 1982–1997 (Children born in Sweden after in vitro fertilization, 1982–1997). Stockholm, Sweden: National Board of Health and Welfare, 2000.
5. Källén B. Search for teratogenic risks with the aid of malformation registries. Teratology 1987;35:47–52.[Medline]
6. Miettinen OS. Simple interval estimation of risk ratio. Am J Epidemiol 1974;100:515–6.
7. Hack M, Brish M, Serr DM, Insler V, Salomy M, Lunenfeld B. Outcome of pregnancy after induced ovulation. Follow-up of pregnancies and children born after clomiphene therapy. JAMA 1970;220:1329–33.
8. Asch H, Greenblatt RB. Update on the safety and efficiency of clomiphene citrate as a therapeutic agent. J Reprod Med 1976;17:175–80.[Medline]
9. Ahlgren M, Källén B, Rannevik G. Outcome of pregnancy after clomiphene therapy. Acta Obstet Gynecol Scand 1976;55:371–5.[Medline]
10. MacGregor AH, Schardein JL. Congenital anomalies and hormones during pregnancy. A critical review. Teratology 1980;7:199–202.
11. Schardein JL. Chemically induced birth defects. New York: Marcel Dekker, 1985:300.
12. Van Loon K, Besseghir K, Eshkol A. Neural tube defects after infertility treatment: A review. Fertil Steril 1992;58: 875–84.[Medline]
13. Werler M, Louik C, Shapiro S, Mitchell AA. Ovulation induction and risk of neural tube defects. Lancet 1994;344: 445–6.[Medline]
14. Greenland S, Ackerman DL. Clomiphene citrate and neural tube defects: A pooled analysis of controlled epidemiological studies and recommendations for future studies. Fertil Steril 1995;64;936–41.[Medline]
15. Ericson A, Källén B, Åberg A. Use of multivitamins and folic acid in early pregnancy and multiple births in Sweden. Twin Res 2001;4:63–6.[Medline]
16. Lynch A, McDuffie R, Murphy J, Faber K, Leff M, Orleans M. Assisted reproductive interventions and multiple birth. Obstet Gynecol 2001;97:195–200.
17. DaCosta ALE, Abdelmassih S, DeOliveira FG, Abdelmassih V, Abdelmassih R, Nagy ZP, et al. Monozygotic twins and transfer at the blastocyst stage after ICSI. Hum Reprod 2001;16:333–6.
18. Schachter M, Raziel A, Friedler S, Strassburger D, Bern O, Ronel R. Monozygotic twinning after assisted reproductive techniques: A phenomenon independent of micromanipulation. Hum Reprod 2001;16:1264–9.
This article has been cited by other articles:
|
|
 |
|
  K I Aston, C M Peterson, and D T Carrell Monozygotic twinning associated with assisted reproductive technologies: a review Reproduction, October 1, 2008; 136(4): 377 - 386. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Hansen, L. Colvin, B. Petterson, J. J. Kurinczuk, N. de Klerk, and C. Bower Admission to hospital of singleton children born following assisted reproductive technology (ART) Hum. Reprod., June 1, 2008; 23(6): 1297 - 1305. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Sun, M. Vestergaard, J. Christensen, J. L. Zhu, B. H. Bech, and J. Olsen Epilepsy and febrile seizures in children of treated and untreated subfertile couples Hum. Reprod., January 1, 2007; 22(1): 215 - 220. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kapiteijn, C.S. de Bruijn, E. de Boer, A.J.M. de Craen, C.W. Burger, F.E. van Leeuwen, and F.M. Helmerhorst Does subfertility explain the risk of poor perinatal outcome after IVF and ovarian hyperstimulation? Hum. Reprod., December 1, 2006; 21(12): 3228 - 3234. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R J Berry, R Kihlberg, and O Devine Impact of misclassification of in vitro fertilisation in studies of folic acid and twinning: modelling using population based Swedish vital records BMJ, April 9, 2005; 330(7495): 815. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Pinborg, A. Loft, L. Schmidt, G. Greisen, S. Rasmussen, and A. N. Andersen Neurological sequelae in twins born after assisted conception: controlled national cohort study BMJ, August 7, 2004; 329(7461): 311. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. S. Green Risks of Birth Defects and Other Adverse Outcomes Associated With Assisted Reproductive Technology Pediatrics, July 1, 2004; 114(1): 256 - 259. [Full Text] [PDF]
|
|
|
|
|
|
A. Pinborg, A. Loft, S. Rasmussen, L. Schmidt, J. Langhoff-Roos, G. Greisen, and A. Nyboe Andersen Neonatal outcome in a Danish national cohort of 3438 IVF/ICSI and 10 362 non-IVF/ICSI twins born between 1995 and 2000 Hum. Reprod., February 1, 2004; 19(2): 435 - 441. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
