1 The Fertility Clinic, Rigshospitalet, 2100 Copenhagen Ø, 2 The National Board of Health, 4th Office, Amaliegade 13, Postbox 2020, 1012 Copenhagen K and 3 The Danish Fertility Society, 4 The Fertility Clinic, Odense University Hospital, 5000 Odense C, Denmark
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: assisted reproduction/congenital malformation/fetal karyotype/National Registry/perinatal data
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The Danish National IVF Registry was established on January 1, 1994. This registry has a number of advantages compared to other registries on assisted reproductive technology. First of all, it is a compulsory registry including every assisted reproductive technology cycle performed in Denmark both in private and public clinics. Secondly, each person is identified by her personal identification number (Central Person Registry, CPR number). This allows each person to be identified in any Danish Medical registry. The data on assisted reproductive technology treatments can therefore be linked to data from other registries.
The present report gives data from the Danish IVF Registry with special focus on births after IVF and intracytoplasmic sperm injection (ICSI). Based on cross-linking data to other registries the study includes data on malformations and use of cytogenetic prenatal diagnosis. Data were compared to a control group with similar maternal age, parity, multiplicity and year of birth.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The records include a number of clinical and laboratory variables in relation to each cycle. After completion of each cycle the corresponding record sheets were sent to the National Board of Health. Each woman was identified by her name and CPR number. Each record was signed by the doctor responsible for the treatment and each clinic had an identification number. During the period there were nine private and six public clinics offering assisted reproductive technology.
Data regarding the treatment were recorded on a treatment form and if pregnancy was obtained also on an `pregnancy outcome form'. The main recordings included the indications, number of oocytes, the methods of fertilization like IVF or ICSI and the number of embryos transferred. Data regarding frozen embryo replacement (FER) and egg donation (ED) were recorded likewise.
The `pregnancy outcome form' included data on misfalls: spontaneous or induced abortions, ectopic pregnancies and fetal reduction. Data on births included: number of infants delivered, sex, gestational age (GA), birthweight and mode of delivery.
Data from other registries
The data from the IVF registry were linked with data from January 1, 1994 to October 31, 1997 from the following registries:
For all registries, the International Classification of Diseases and Health (ICD-10) (1997) was used when coding the hospital contact.
Control group
Data on pregnancy and childbirth after assisted reproductive technology were compared to a control group of non-assisted reproductive technology treated women and their children. The control group was generated through registry data from the Danish Medical Birth Registry. The control group was matched using random selection within strata by maternal year of age, by child year of age, exact matching for parity and finally matching for multiple pregnancies (1756 women delivering 2245 children).
Furthermore, data were compared to all pregnancies and childbirths in Denmark in 1995 (Medical Birth and Malformation Statistics 1994Medical Birth and Malformation Statistics 1995; Danish National Board of Health, 1997a).
It was not possible to compare data on prenatal cytogenetic findings to a control group, as the number of clinical pregnancies in the control group was unknown. Instead, these data were compared to data on all prenatal cytogenetic findings in Denmark in 1995 (Cytogenetic Central Registry, 1997).
Definitions
Gestational age (GA) for both study and control group were found through registry data. It was assumed that for the study group the calculations were based on the day of embryo transfer.
Prematurity was defined as delivery before 37 weeks of GA. Low birthweight (LBW) was defined as <2500 g at birth and very low birthweight (VLBW) as <1500 g. Stillbirth was defined as intrauterine death after 27 completed weeks of GA.
Statistics
Statistical analyses were based on the 2 and MannWhitney tests. Statistical significance was defined as a probability value of P < 0.05.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
There was a total of 1756 births resulting in 2245 children. Of these 1.3% were stillborn. Of 2245 children 1913 (85.2%) were after IVF, 180 (8.0%) were after ICSI, 105 (4.7%) were after FER and 47 (2.1%) were after ED (Table I).
In the control group, 0.8% of the children were stillborn.
Of all deliveries, 1298 were singleton (73.9%), 427 were twin (24.3%) and 31 were triplet deliveries (1.8%). The mortality in relation to multiplicity is shown in Table II.
|
Prenatal diagnosis
Among assisted reproductive technology-treated women, 289 had prenatal genetic examinations (PGE). Of these 78.2% were amniocentesis and 21.8% chorion villus biopsies. This accounted for 3.8% of all PGE in Denmark in 1995. In total, 13.2% of all assisted reproductive technology clinical pregnancies were examined, or 15.4% of all pregnancies ending in delivery. An abnormal fetal karyotype was found in 10 (3.5%) of 289 assisted reproductive technology pregnancies tested (Table III).
|
There were three (5.4%) chromosomal abnormalities among women who had PGE after ICSI compared to seven (3.4%) after IVF (Table III). Of the assisted reproductive technology-treated women aged
35 years, only 34.6% were examined. Of all women who gave birth in 1995 and had a PGE, 57.6% were
35 years.
Malformations in livebirths and stillbirths
Of 2245 children born in the study group, 107 (4.8%) had malformations. Of these, 45.8% were born in multiple pregnancies.
In the control group, 103 children (4.6%) were born with malformations (NS). Of these, 60.2% were born in multiple pregnancies. Of all children born in Denmark in 1995, 1987 (2.8%) had malformations.
The types of major malformations found at birth in the study and control group are listed in Table IV. In total, five children in the study group with malformations were stillborn. All five were after IVF treatment. One singleton died in utero with cervical spina bifida and unilateral cleft palate. One twin with trisomy 18 died in utero, one twin with LaurenceMoonBiedlBardet's syndrome died in utero. One twin died in utero with a defect in the ventricular septum. One twin was anencephalic and died in utero.
|
Neonatal data
In total, 23.8% of the children in the study group were delivered preterm compared to 23.4% in the control group (NS) (Figure 1).
|
Figure 2 shows the birthweight distribution in the study and control group for singleton, twin and triplet deliveries. In total, 23.6% of the assisted reproductive technology children were LBW (singletons 7%, twins 42.2% and triplets 87.1%) compared to 23.8% (NS) in the control group [singleton 4.8% (P < 0.05), twins 44.8% (NS) and triplets 96.8% (NS)]. In the study group, 4.2% were VLBW (singletons 1.5%, twins 6.7% and triplets 20.4%) compared to 5.4% (NS) in the control group [singletons 0.4% (P < 0.005), twins 10.2% (NS) and triplets 32.3% (NS)].
|
For singletons, the mean birthweight was 3389 g in the study group and 3454 g in the control group (P < 0.01). For twins, the mean birthweight was 2516 g in the study group and 2443 g in the control group (P < 0.05). For triplets, the corresponding figures were 1845 g and 1712 g (NS).
Mode of delivery
The rate of spontaneous deliveries for assisted reproductive technology singletons was 27.4% and for multiples 11.4% compared to 34.8% (P < 0.05) and 10.7% (NS) in the control group.
The Caesarean section (CS) rate for assisted reproductive technology singleton deliveries was 13.7% (7.5% elective and 6.2% acute procedures). For multiple pregnancies the corresponding figures were 37.3, 26.2 and 11.1%. In the control group, the corresponding rates of CS in singleton deliveries were: 9.5% (P < 0.05), 5.2% (P < 0.05) and 4.3% (P < 0.05) and in multiple pregnancies: 29.5% (P < 0.05), 20.1% (P < 0.05) and 9.4% (NS).
The CS rate in the background population in Denmark was 12.6% in 1995.
Labour was induced in 4.7% of assisted reproductive technology singleton deliveries and 10.5% of multiple deliveries. In the control group, the corresponding rates were 3.2% for singleton deliveries (P = 0.05) and 11.6% for multiple deliveries (NS).
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In order to estimate the true risks for births after assisted reproductive technology, previous studies used control groups stratified for age and multiple births (Gissler et al., 1997; HFEA, 1997; Socialstyrelsen, 1998) or used control groups of patients who conceived while on the waiting list (Saunders et al., 1988
). We have been able to include parity and child's year of birth in addition to maternal age and multiplicity. Using this control group we found no major differences in relation to GA, birthweight or rate of malformations.
In total, the average birthweight was only 4 g different between the study and control group. This indicates that it is the characteristics of the patients rather than assisted reproductive technology that determines the risks of assisted reproductive technology pregnancies compared to the background population.
Recent Swedish national data (Socialstyrelsen, 1998) have a comparable match and present similar results in relation to birthweight and rates of malformations. The two populations are very similar and the clinical approach likewise, making the data comparable.
Both the Swedish (Socialstyrelsen, 1998) and the present study used the same definitions for the coding of malformations, namely the international ICD-10 (International Classification of Diseases and Health Related Problems, Danish National Board of Health, 1997b
). Use of registry data to define malformations limits the inherent bias of other studies of assisted reproductive technology children, as recently discussed (Simpson, 1996
; Mitchell, 1997
).
There is clearly one methodological limitation in the present study, because we were unable to define the number of women in the control group of non-assisted reproductive technology treated women that conceived following other types of infertility treatment.
The risk of malformations (major and minor) was 4.8% among assisted reproductive technology children. Data derived from the same registry (Danish National Board of Health, 1997a) showed that in the background population the risk of malformations was only 2.8%, and thus much lower than in the present study.
However, looking at the age-matched control group with similar parity and rate of multiple births the rate of malformations was 4.6%, identical to infants born after assisted reproductive technology (IVF and ICSI). Looking at other studies of malformations after assisted reproductive technology, the rates of children born with all types of malformations in our study (4.8%) are comparable, although slightly higher than in other studies. The Swedish data (Socialstyrelsen, 1998) showed that 3.3% of the children were born with major malformations, UK data (Rizk et al., 1991
) likewise 3.0%, Australian data (Lancaster et al., 1997
) 2.6% major malformations after 16 weeks of GA and data from French National Registry (FIVNAT, 1997a
) and other French sources (Rufat et al., 1994
) found a malformation rate of 2.5% of liveborn children and 3.2% when including malformations detected after 12 weeks of GA resulting in an induced abortion or delivery.
Some authors have suggested an increased risk of malformations after ICSI (Kurinczuk and Bower, 1997). In our material, the malformation rate after ICSI was 1.7%. However, the data on ICSI represent a small sample size, as the Danish ICSI programmes had only just started in 1994. Still, our data support most other studies showing that there is no increased risk of malformation among children born after ICSI (Bonduelle et al., 1995
, 1996
, 1998
; Palermo et al., 1996
; Bonduelle and Devroey, 1997
; Mitchell, 1997
).
Of 1756 births, 26.1% were multiple, 24.3% twin deliveries and only 1.8% triplet deliveries (Table III). The twinning rates are almost identical to national data from France, Sweden and Finland (24.525.1%) (FIVNAT, 1997b
; Gissler et al., 1997; Socialstyrelsen, 1998
). However, the Danish triplet rates were lower than in the French, Swedish and English registries (2.63.5%) (Tan et al., 1992
; FIVNAT, 1997; Socialstyrelsen, 1998
). In 19941995, Danish assisted reproductive technology centres transferred two or three embryos. Despite the low triplet rates due to a restricted number of embryos transferred, the delivery rate in IVF was 20% per initiated stimulation cycle. From 1997 onward, a maximum of two embryos are transferred. There were no quadruplets in 19941995. Recent data from Sweden (Karlström et al., 1998
) confirm our data; they found a delivery rate per embryo transfer of 25.1% despite a decrease of transfers with three pre-embryos from 77 to 7%. At the same time, they found a reduction in the multiple pregnancy rate from 34 to 25%.
The Danish rates of preterm deliveries were 7.3% for singletons, 41.2% for twins and 93.5% for triplets. These rates seem to be the lowest rates ever reported. Compared to the appropriate control group, the rates were basically similar (although significantly higher among singletons), but different from Danish national data and previous studies. The significant difference between singletons could be due to a much higher rate of assisted reproductive technology singletons born before 28 weeks of GA; however, the numbers are small and this difference has to be analysed in the following years, as the numbers increase. Danish legislation define stillborn as born after 27 weeks of GA; before that the fetus is defined as an abortion and furthermore, the parents have no rights to parental leave.
National data from Sweden (Socialstyrelsen, 1998) reported preterm deliveries among 11.2% of singletons and 47.3% among twins. The corresponding figures in the Finnish study (Gissler et al., 1997) were 9% and 61% for multiples. Data from Australia (Lancaster et al., 1997
) reported likewise 12.8% for singletons and 55.9% for twins. French data showed preterm rates of 9.1 and 42.4% (FIVNAT, 1997b
). Regional data from the United States (Tallo et al., 1995
) and a controlled study from Israel (Moise et al., 1998
) showed significant differences between IVF twins and control twins.
The percentage of children who were LBW was 23.6% in the study group (singletons 7%, twins 42% and triplets 87%) and comparable with the control group 23.8% (singleton 4.8%, twins 44.8% and triplets 96.8%).
The percentage of children who were VLBW was 4.2% in the study group (singletons 1.5%, twins 6.7% and triplets 20.4%) compared to 5.4% in the control group (singletons 0.4, twins 10.2 and triplets 32.3%). Data from the UK (Beral et al., 1990) showed much higher rates in children born after assisted reproductive technology (in total 32% LBW, 12% of singletons, 55% of twins and 94% of triplets) and 6% VLBW. One possible explanation for this difference may be because the UK data were from 1978 to 1987, where the obstetrical management of assisted reproductive technology pregnancies may have been different from today. Data from France (FIVNAT, 1997b) reported 30.6% of all children were LBW (singleton 9.5% and twins 53.9%). Recent data from Australia (Gibson et al., 1998
) showed significantly more LBW IVF singletons compared to controls and data from Israel showed significantly more VLBW IVF twins compared to controls (Moise et al., 1998
).
Of the 5219 women treated with assisted reproductive technology reported in our material, 2186 had a clinical pregnancy. Of the clinical pregnancies, only 13.2% had a PGE and 3.5% of these had an abnormal karyotype; all but one was associated with a maternal age of 35 years.
Unfortunately, the methodology used in this present study did not allow us to compare the examined women to a control group of pregnant women equal in age, parity and number of children in the pregnancy. The reason is that we had to define our control group among 1756 matched women who actually delivered a child. We can therefore only compare to the background population in Denmark.
The in-vitro treated women have a higher age and therefore a higher risk of abnormal karyotype (Snijders and Nicolaides, 1996). They are more often nulliparous and some deliver after ICSI, which in some studies have shown a higher incidence of chromosomal abnormalities (Kurinczuk et al., 1997). However, our data show that the patients are reluctant to use PGE.
Data from Finland (Gissler et al., 1997) and the UK (Beral and Doyle, 1990) showed that 10% of the treated women had a PGE. Only data from Belgium (Bonduelle et al., 1998
) have shown a higher incidence of PGEs (40%). However, these women were all invited to take part in a special study of ICSI and all were offered prenatal genetic counselling.
Data from Germany (Wurfel et al., 1992) showed a 7.3% risk of abnormal karyotypes. Of these, two were clinically significant (trisomy 21). Some studies have shown a higher incidence of abnormal karyotypes after ICSI (Kurinczuk et al., 1997) compared to IVF. Our data confirm data from Belgium (Bonduelle et al., 1995
) where no differences were found. However compared to other studies of PGEs the number of abnormal karyotypes for ICSI treated in our material was higher (5.4%). In the Belgian material 1.2% had abnormal karyotypes and 1.2% had familial structural abnormalities (Bonduelle et al., 1996
).
One must take into account that our sample size is fairly small as most Danish clinics began their ICSI programmes around 1995.
Finally, we do not in the Danish Registry data have any information on the karyotype of the involved parents, and have not been able to control the karyotype of the children after birth.
The rate of CS was significantly different between the study and control groups. This could indicate that when handling delivery of assisted reproductive technology children the Danish obstetricians take more precautions, because these pregnancies are believed to be more vulnerable. However, Danish rates were lower compared to French assisted reproductive technology rates (FIVNAT, 1997b) where in total, 25.7% of the singletons (15.3% elective) and 54.6% of the twins (37.6% elective) were delivered by CS.
In conclusion, comparing data from assisted reproductive technology to a control group matched on both maternal age, parity, multiplicity and child's year of birth, our data indicate that it is the characteristics of the patient and multiplicity of pregnancy, rather than the assisted reproductive technology, that determine the higher risk of preterm birth, low birth weight and malformations of these pregnancies compared to the background population. Data from the Danish IVF Registry will be followed in the years to come.
![]() |
Acknowledgments |
---|
![]() |
Notes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bonduelle, M. and Devroey, P. (1997) Commentary: major defects are overestimated. Br. Med. J., 315, 12651266.
Bonduelle, M., Legien, J., Derde, M.P. et al. (1995) Comparative follow-up study of 130 children born after intracytoplasmic sperm injection and 130 children born after in-vitro fertilization. Hum. Reprod., 10, 33273331.[Abstract]
Bonduelle, M., Willikens, J., Buysse, A. et al. (1996) Prospective study of 887 children after intracytoplasmatic sperm injection with ejaculated epididymal and testicular spermatozoa and after replacement of cryopreserved embryos obtained after ICSI. Hum. Reprod., 11, 131159.[ISI][Medline]
Bonduelle, M., Wilikens, A., Buysse, A. et al. (1998) A follow-up study of children born after intracytoplasmic sperm injection (ICSI) with epididymal and testicular spermatozoa and after replacement of cryopreserved embryos obtained after ICSI. Hum. Reprod., 13 (Suppl. 1), 196207.[Medline]
Cytogenetic Central Registry (1997) Prenatal Genetic Examinations in Denmark in 1995. Århus Kommunehospital. Århus, Denmark.
Danish National Board of Health (1997a) Danish Medical Birth and Malformation Statistics 1994 and 1995. Sundhedsstatistikken 1997:6. Munksgaard. Copenhagen, Denmark.
Danish National Board of Health (1997b) International Classification of Diseases and Health Related Problems. 10. Revision. 1993. Munksgaard. Copenhagen, Denmark.
D.I.R. (1997) Deutsches IVF Register. Jahrbuch 1997. D.I.R. Bundesgeschäfsstelle Ärztekammer Schleswig-Holstein, Bismarckallee 812. 23795 Bad Segeberg, Germany.
FIVNAT 1996 report (1997a) French National Register on In vitro Fertilisation. Contracept. Fertil. Sex., 25, 499502.[Medline]
FIVNAT (1997b) Bilan de Lánnee 96. Dossier FIVNAT 97.
Gibson, F.L., Ungerer, J.A., Leslie, G.I. et al. (1998) Development, behaviour and temperament: a prospective study of infants conceived through in-vitro fertilization. Hum. Reprod., 13, 17271732.[Abstract]
Gissler, M. and Tiitinen, A. (1997) IVF-Statistics in Finland 19921996. 12/97. STAKES, National Research and Development Centre for Welfare and Health. Helsinki, Finland.
Human Fertilisation and Embryology Authority (HFEA) (1997) The Patients' Guide to DI and IVF Clinics. HFEA, Paxton House, 30 Artillery Lane, London E1 7LS, UK.
Karlström, P.O., Nygren, K.G., Wikland, M. et al. (1998) Does a shift to two embryo transfers after in vitro fertilization decrease the success and multiple birth rate? An analysis of assisted reproductive technologies in Sweden 19911996. 16th World Congress on Fertility and Sterility. Fertil. Steril., S351, P-711.
Kurinczuk, J.J. and Bower, C. (1997) Birth defects in infants conceived by intracytoplasmatic sperm injection: an alternative interpretation. Br. Med. J., 315, 12601265.
Lancaster, P., Shafir, E., Hurst, T. et al. (1997) Assisted Conception, Australia and New Zealand. 1994 and 1995. AIHW National Perinatal Statistics Unit (Assisted Conception Series: no. 2), Sydney.
Mitchell, A.A. (1997) Intracytoplasmatic sperm injection: offering hope for term pregnancy and a healthy child? Br. Med. J., 315, 12451246.
Moise, J., Laor, A, Armon, Y, et al. (1998) The outcome of twin pregnancies after IVF. Hum. Reprod., 13, 17021705.[Abstract]
Palermo, G.D., Colombero, L.T., Schattman, G.L. et al. (1996) Evolution of pregnancies and initial follow-up of newborns delivered after intracytoplasmatic sperm injection. JAMA, 276, 18931897.[Abstract]
Rizk, B., Doyle, P., Tan, S.L. et al. (1991) Perinatal outcome and congenital malformations in in-vitro fertilization babies from the BournHallam group. Hum. Reprod., 6, 12591264.[Abstract]
Rufat, P., Olivennes, F., de Mouzon, J. et al. (1994) Task force report on the outcome of pregnancies and children conceived by in vitro fertilization (France: 19871989). Fertil. Steril., 61, 324330.[ISI][Medline]
Saunders, D.M., Mathews, M. and Lancaster, P. (1988) The Australian Register: current research and future role. Ann. N. Y. Acad. Sci., 541, 712.[ISI][Medline]
Simpson, J.L. (1996) Registration of congenital anomalies in assisted reproductive technology populations: pitfalls. Hum. Reprod., 11, 8188.[Abstract]
Snijders, R.J.M. and Nicolaides, K.H. (1996) Ultrasound markers for fetal chromosomal defects. In Frontiers in Fetal Medicine. The Parthenon Publishing Group, London, UK.
Socialstyrelsen (1998) Forlösningar efter provrörs-befruktningar i Sverige. Socialstyrelsen. Stockholm, Sweden.
Society for Assisted Reproductive Technology Registry (SART) and the American Society for Reproductive Medicine (ASRM) (1998) Assisted reproductive technology in the United States and Canada: 1995 results generated from the American Society for Reproductive Medicine/Society for Assisted Reproductive Technology Registry. Fertil. Steril., 69, 389398.[ISI][Medline]
Tallo, C.P., Vohr, B., Oh, W. et al. (1995) Maternal and neonatal morbidity associated with in vitro fertilization. J. Pediat., 127, 794800.[ISI][Medline]
Tan, S.L., Doyle, P., Campbell, S. et al. (1992) Obstetric outcome of in vitro fertilization pregnancies compared to normally conceived pregnancies. Am. J. Obstet. Gynecol., 167, 778784.[ISI][Medline]
Templeton, A., Morris, J.K., and Parslow, W. (1996) Factors that affect outcome of in-vitro fertilisation treatment. Lancet, 348, 14021407.[ISI][Medline]
World Collaborative Report on In vitro Fertilisation (1997) Preliminary data from 1995. International Working Group for Registers on Assisted Reproduction. J. Assist. Reprod. Genet., 14, 251264.
Wurfel, W., Haas-Andela, H., Krusmann, G. et al. (1992) Prenatal diagnosis by amniocentesis in 82 pregnancies after in vitro fertilisation. Eur. J. Obstet. Gynecol. Reprod. Biol., 44, 4752.[ISI][Medline]
Submitted on November 2, 1998; accepted on March 9, 1999.