RBMOnline - Vol 11. No 1. 2005 76 85 Reproductive BioMedicine Online; www.rbmonline.com/article//1705 on web 2 June 2005 Article Obstetric and perinatal outcome of 1655 ICSI and 3974 IVF singleton and 1102 ICSI and 2901 IVF twin births: a comparative analysis Willem Ombelet started his career in infertility and IVF in 1984 in Pretoria, South Africa. Since 1987 he has been working in the Department of Obstetrics and Gynaecology in the St Jans Hospital of Genk, Belgium. He is the founder of the Genk Institute for Fertility Technology and Chairman of four Andrology in the Nineties meetings. In 1998 he obtained his PhD degree at the University of Leuven. His thesis was entitled The value of sperm morphology and other semen parameters in diagnosis and treatment of human subfertility. In 1999 he became Head of the Department, and from 2001 to 2004 he was the President of the Flemish Society of Obstetrics and Gynaecology. Dr Willem Ombelet Willem Ombelet 1,3,5, Isabelle Cadron 3, Jan Gerris 1, Petra De Sutter 1, Eugene Bosmans 3, Guy Martens 2, Gunther Ruyssinck 2, Paul Defoort 1,2, Geert Molenberghs 4, Wilfried Gyselaers 3 1 Scientific Board of the Flemish Society of Obstetrics and Gynecology; 2 SPE (Studiecentrum voor Perinatale Epidemiologie), Brussels; 3 Genk Institute for Fertility Technology, Department of Obstetrics and Gynecology, Genk; 4 Centre for Statistics, Limburgs Universitair Centrum, Diepenbeek, Belgium 5 Correspondence: e-mail: willem.ombelet@pandora.be Abstract A total of 3974 IVF and 1655 ICSI singleton births and 2901 IVF and 1102 ICSI twin births were evaluated. Pregnancies after both fresh and frozen transfers were included. IVF and ICSI singleton pregnancies were very similar for most obstetric and perinatal variables. The only significant difference was a higher risk for prematurity (<37 weeks of amenorrhoea) in IVF pregnancies compared with ICSI pregnancies (12.4 versus 9.2%, OR = 1.39, 95% CI = 1.15 1.70). For twin pregnancies, differences were not statistically different except for a higher incidence of stillbirths in the ICSI group (2.08 versus 1.03%, OR = 2.04, 95% CI = 1.14 3.64). Intrauterine growth retardation with or without pregnancy-induced hypertension was observed more often in the ICSI group. Regression analysis of the data with correction for parity and female age showed similar results for twins. For singletons, this analysis showed similar results with the exception of low birth weight babies (<2500 g), which were also observed more often in IVF pregnancies (9.6 versus 7.9%, OR = 0.79, CI = 0.65 0.98, P = 0.03). This large case-comparative retrospective analysis showed that the obstetric outcome and perinatal health of IVF and ICSI pregnancies is comparable. Keywords: human, ICSI, IVF, obstetric outcome, perinatal outcome, pregnancy 76 Introduction Several studies have reported on the perinatal outcome of pregnancies obtained after IVF and intracytoplasmic sperm injection (ICSI). The first reports from registries with retrospectively collected data from IVF pregnancies were reassuring (National Perinatal Statistics Unit and Fertility Society of Australia, 1988; Beral et al., 1990). A few reports mentioned a statistically relevant increase in major birth defects when IVF and ICSI pregnancies were compared with the general population or a matched group of naturally conceived gestations (Bergh et al., 1999; Wennerholm et al., 2000; Hansen et al., 2002; Koivurova et al., 2002a; Strömberg et al., 2002), while other studies did not show any difference at all (Ericson and Källén, 2001; Anthony et al., 2002). Nowadays the opinion is widely held that the perinatal outcome of pregnancies resulting from assisted reproduction is substantially worse than that of pregnancies following natural conception. This is mainly attributed to a higher rate of multiple pregnancies, which in turn is associated with a higher rate of perinatal mortality and morbidity (Luke and Keith, 1992; Gissler et al., 1995; ESHRE Capri Workshop Group, 2000, 2003; Blickstein, 2003). However, several studies have suggested that even singleton pregnancies after IVF and ICSI carry a higher risk for low birth weight and prematurity when compared with naturally conceived singletons (Tan et al., 1992; Tanbo et al., 1995; Verlaenen et al., 1995; Reubinoff et al., 1997; Dhont et al. 1999; Koudstaal et al., 2000; Koivurova et al., 2002a,b;
Schieve et al., 2002, Helmerhorst et al., 2004; Jackson et al., 2004; McGovern et al., 2004). The reason why perinatal health problems occur more frequently in assisted reproduction pregnancies is still unknown, but could be related to different factors such as invitro technology, the medication used in assisted reproduction programmes, or the infertility itself (Lambert, 2003). With the introduction of ICSI in 1992 (Palermo et al., 1992), even severe male subfertility could be treated successfully. From the very beginning, ICSI was considered a more risky procedure. From the point of view of perinatal health, there are major differences between IVF and ICSI in both patient selection and the techniques themselves. Two types of risks are to be considered: ICSI procedure-dependent and ICSI procedure-independent risk factors (Patrizio, 1995; Bonduelle, 2003). Procedure dependent risks are caused by the procedure itself, involving the penetration of the zona pellucida and oocyte cytoplasm. Procedure-independent risks are related to the cause of infertility, leading to the use of spermatozoa that normally cannot achieve fertilization. Ludwig and Kataliniç (2003) showed that the course of pregnancy as well as the outcome after ICSI is not affected by the origin of spermatozoa and by the number of sperm in the ejaculate. On the other hand, in a prospective, well controlled, nationwide German cohort study of more than 3000 ICSI births, it was shown that the major congenital malformation rate was increased after ICSI compared with natural conception (Kataliniç et al., 2004). Obviously even more questions are raised when epididymal or testicular spermatozoa are used in men with non-obstructive azoospermia. To investigate the possible risks of IVF and ICSI on perinatal health, a case-comparative analysis was conducted to compare the perinatal outcome of all IVF and ICSI singleton and twin pregnancies in the Dutch-speaking part of Belgium (Flanders) in the period from January 1997 until December 2003. Materials and methods A retrospective study was performed by analysing the data from the Studiecentrum voor Perinatale Epidemiologie (SPE). The SPE collects data on the medical and obstetric history, and on perinatal events of each hospital delivery in Flanders of >21 weeks of gestational age or 500 g birth weight. Full co-operation of all 80 obstetrics departments in Flanders (52 55% of all deliveries in Belgium) has been established since 1991. The SPE also gathers data on a substantial percentage (70%) of the Flemish home births, which, however, number only 600 each year. The data are based on questionnaires completed by midwives, obstetricians and paediatricians in the early neonatal period. The obstetric and perinatal file registers 33 items of data per child. If the newborn is transferred to the neonatology unit, data for another 20 neonatal items are recorded. All data are sent to a data coordinator for review of errors and omissions. Correction and completion is ensured by telephone calls, additional questionnaires and, if necessary, visits to local departments. Subsequently the files are stored in a computer database. Each year, a complete analysis of the data is performed. This results in a yearly global report for the region and an individual report provided to every obstetric unit of its site-specific outcome. From 1997 onwards, a distinction was made between IVF and ICSI pregnancies. These data include pregnancies after fresh and frozen embryo transfers following IVF and ICSI procedures. Out of 430,565 births and 422,441 deliveries between 1 January 1997 and 31 December 2003, 1655 ICSI and 3974 IVF singleton and 1102 ICSI and 2901 IVF births were selected. The main outcome measures for this study were gestational age, birth weight, fetal presentation at delivery, mode of delivery, induction of labour, fetal sex, Apgar score <5 after 5 min, admission to a neonatal intensive care unit (NICU), congenital malformations, perinatal mortality and perinatal morbidity including intracranial bleeding, convulsions, need for assisted ventilation and respiratory distress syndrome. For twin deliveries, birth weight discordancy was also included. A distinction was made between normal birth weight (>2500 g), low birth weight (LBW, <2500 g) and very low birth weight (VLBW, <1500 g). With respect to gestational age, a distinction was made between term birth ( 37 weeks), preterm (<37 weeks) and extreme preterm (<32 weeks) birth. Birth weight discordance was the weight difference between the pair of twins, expressed as the percentage of the birth weight of the largest child. Three categories were made: birth weight discordance <15%, between 15 and 25% and 25%. Stillbirth is defined as the birth of a lifeless child >500 g, neonatal death is the death of a live born child >500 g within 7 days after birth. The perinatal mortality rate is the sum of stillbirths and neonatal deaths divided by the total number of live and stillbirths. Only minor and major congenital malformations, recognized during pregnancy or during hospitalization in the neonatal period, were reported. Minor and major malformations were not differentiated. Statistics The Mann Whitney U-test, a classical non-parametric test for group comparison (Lehman, 1975), was used to compare parity, duration of pregnancy, birth weight and age of the mother between IVF and ICSI singleton pregnancies and between IVF and ICSI twin pregnancies. For the statistical analysis comparing obstetric and perinatal outcome differences between IVF and ICSI pregnancies, the odds ratio, 95% confidence interval and its corresponding P-values were used. A difference at the 5% level of significance was considered the threshold of probability. The significance of the difference between ICSI and IVF in singleton and twin pregnancies was also investigated after correction for possible effects of parity and age of the mother. Multiple logistic regression models were used for this purpose. The model included an effect of the technique (IVF or ICSI), female age and parity. The effects of both parity and age on the logit scale were considered to be linear. Results Out of 430,565 births and 422,441 deliveries, 2757 (0.64%) ICSI and 6875 (1.60%) IVF births were found. 77
Singleton pregnancies Data on pregnancy and perinatal outcome of 1655 ICSI and 3975 IVF singleton pregnancies are summarized in Tables 1 and 2. Gestational age (P < 0.001), parity (P < 0.001), birth weight (P = 0.045) and maternal age (P = 0.002) were significantly different between both groups. For maternal age and parity, these differences were small and probably do not have major clinical importance. A significant lower mean duration of pregnancy and mean birth weight in the IVF group can be explained by a higher incidence of prematurity (<37 weeks), which was observed more frequently in the IVF group (12.4 versus 9.2%, OR = 1.39, CI = 1.15 1.70). The incidence of severe prematurity (<32 weeks) and very low birth weight was similar in both groups. Logistic regression analysis after adjustment for parity and female age showed similar results, with the exception of low birth weight, which was significantly increased in IVF compared with ICSI pregnancies (OR = 0.79, CI = 0.65 0.98, P = 0.03). The Caesarean section rate for singleton pregnancies was 22.3 and 23.8% for ICSI and IVF subjects respectively. This difference was not significant. There was no significant difference for any of the other major outcome measures. Congenital malformations were observed in 2.2% of ICSI babies compared with 2.3% for IVF babies (difference not significant). Twin pregnancies Parity was significantly different between both groups (P = 0.008, 1.4 for ICSI, 1.5 for IVF pregnancies), but this difference seems to have no major clinical importance. As shown in Tables 3 and 4, there was no difference between the ICSI (1102 births) and IVF group (2901 births), with the exception of a significantly higher stillbirth rate in the ICSI group (2.08 versus 1.03%; OR = 2.04, CI = 1.14 3.64). Logistic regression analysis with correction for parity and female age showed similar results. (stillbirth: OR = 1.99, CI = 1.15 3.44, P = 0.01). A detailed study of all stillbirths (Tables 5 and 6) revealed a very high incidence of cervical incompetence with or without preterm premature rupture of membranes in 16 out of 30 IVF births (53.3%), versus 11 out of 23 for ICSI (47.8%). The rate of cervical incompetence was higher in ICSI compared with IVF pregnancies (11/551 or 2.0% for ICSI versus 16/1451 or 1.1% for IVF), although this difference was not significant. Intrauterine growth retardation with or without pregnancy-induced hypertension was observed more often in the ICSI group(9/23 or 39.1% versus 4/30 or 13.3% for IVF). The birth weight discordance was not significantly different between both groups (birth weight discordance 25%: 15.8% for IVF versus 15.4% for ICSI, birth weight discordance 15 25%: 22.1% for IVF versus 19.1% for ICSI, birth weight discordance <15%: 62.2% for IVF versus 58.4% for ICSI). The mean birth weight difference between both members of a twin was 34.9 g for IVF twins and 57.6 g for ICSI twins. This difference was not statistically significant. Discussion Data registration of the obstetrical and perinatal outcome of pregnancies after assisted reproduction is crucial and should be integrated into statutory regulations or national guidelines (Templeton, 2003). According to the literature, pregnancies and births following assisted reproduction are at increased risk for perinatal health problems. The presence of a higher incidence of multiple pregnancies seems to be the most important risk factor and therefore single embryo transfer policies are promoted and introduced in many countries (Hansen et al., 2002; Schieve et al., 2002; De Neubourg and Gerris, 2003; Wennerholm, 2003; Gurgan and Demirol, 2004; Ombelet et al., 2005). However, singleton pregnancies after IVF and ICSI are also reported to carry a higher risk for cerebral palsy, premature birth, low birth weight and multiple birth defects (Hansen et al., 2002; Koivurova et al., 2002a; Schieve et al., 2002; Strömberg et al., 2002). These health problems in assisted reproduction babies seem to originate from the infertility itself and not from other possible causative factors such as ovarian stimulatory drugs and the in-vitro procedure (Lambert, 2003). ICSI was introduced in the early 1990s as a valuable treatment option in case of severe male subfertility (Palermo et al., 1992). ICSI bypasses all natural sperm selection processes and this questions the safety of this technique. One of these was a possibly higher risk of congenital abnormalities in the ICSI children compared with IVF and naturally conceived babies. In the current study, the obstetric and perinatal outcomes of IVF versus ICSI were compared in both singleton and twin pregnancies using the data of the SPE in Flanders, Belgium. Since the obstetric and perinatal outcome is strongly linked to the number of fetuses, a comparative analysis of both groups was made considering singleton and twin pregnancies separately. No major difference was found in obstetrical and/or perinatal outcome between IVF and ICSI pregnancies, with the Table 1. General data for ICSI and IVF singleton births. Data are means ± SD unless otherwise indicated. ICSI IVF P-value a 78 Births (n) 1655 3974 Gestational age (weeks) 38.6 ± 2.2 38.4 ± 2.2 <0.001 Parity 1.4 ± 0.7 1.5 ± 0.8 <0.001 Birth weight (g) 3227 ± 585 3193 ± 610 0.045 Maternal age (years) 32.0 ± 4.2 32.3 ± 4.2 0.002 a Mann Whitney U-test.
Table 2. Comparison of obstetric and perinatal data of 1655 ICSI and 3974 IVF singleton births (OR = odds ratio, CI = 95% confidence interval, GA = gestational age, NICU = neonatal intensive care unit, IC bleeding = intracranial bleeding, RDS = respiratory distress syndrome). ICSI IVF OR (95% CI) P-value a n % n % Obstetric data Caesarean delivery 369 22.3 947 23.8 0.92 (0.80 1.05) NS Induction of labour 513 31.0 1238 31.1 0.99 (0.88 1.22) NS Breech presentation 97 5.9 259 6.5 0.89 (0.70 1.14) NS Perinatal data GA <32 weeks 27 1.6 74 1.9 0.87 (0.56 1.36) NS GA <37 weeks 152 9.2 491 12.4 0.58 (0.47 0.71) <0.001 Birth weight <1500 g 32 1.9 81 2.0 0.95 (0.63 1.43) NS Birth weight <2500 g 132 7.9 381 9.6 0.82 (0.66 1.01) NS Fetal sex male 846 51.1 2031 51.1 1.00 (0.89 1.12) NS Congenital malformations 36 2.1 90 2.4 0.96 (0.65 1.42) NS Apgar score <5 7 0.4 17 0.4 0.99 (0.41 2.39) NS Transfer to NICU 323 19.5 857 21.6 0.88 (0.76 1.02) NS Perinatal death 20 1.20 44 1.11 1.09 (0.62 1.91) NS Stillbirth 13 0.78 31 0.78 1.01 (0.50 2.00) NS Neonatal death 7 0.42 13 0.32 1.29 (0.47 3.47) NS Assisted ventilation 29 1.7 98 2.5 0.71 (0.46 1.07) NS IC bleeding 6 0.4 28 0.7 0.51 (0.21 1.24) NS Convulsions 4 0.2 14 0.3 0.69 (0.23 2.08) NS RDS 11 0.6 45 1.1 0.58 (0.30 1.13) NS a Mann Whitney U-test. NS = not statistically significant. Table 3. General data for ICSI and IVF twin births. Data are means ± SD unless otherwise indicated. ICSI IVF P-value Births (n) 1102 2901 Deliveries (n) 551 1451 Gestational age (weeks) 35.4 ± 3.0 35.0 ± 3.0 NS Parity 1.4 ± 0.6 1.5 ± 0,8 0.008 Birth weight (g) 2317 ± 591 2339 ± 580 NS Maternal age (years) 31.6 ± 3.9 31.6 ± 3.7 NS a Mann Whitney U-test NS = not statistically significant. 79
Table 4. Comparison of obstetric and perinatal data of 1102 ICSI and 2901 IVF twin births (OR = odds ratio, CI = 95% confidence interval, GA = gestational age, NICU = neonatal intensive care unit, IC bleeding = intracranial bleeding, RDS = respiratory distress syndrome). ICSI IVF OR (95% CI) P-value a n % n % Obstetric data Caesarean delivery 263 47.7 749 51.6 0.90 (0.77 1.06) NS Induction of labour 145 26.3 370 25.5 1.05 (0.83 1.32) NS Breech presentation 282 25.6 701 24.1 1.08 (0.92 1.27) NS Perinatal data GA <32 weeks 48 8.8 124 8.5 1.02 (0.73 1.43) NS GA <37 weeks 319 58.2 828 57.0 1.05 (0.86 1.29) NS Birth weight <1500 g 99 9.0 248 8.5 1.06 (0.83 1.35 NS Birth weight <2500 g 652 59.2 1675 57.7 1.06 (0.92 1.22) NS Fetal sex male 549 49.8 1448 49.9 0.99 (0.87 1.14) NS Congenital malformations 35 3.2 82 2.8 1.13 (0.74 1.71) NS Apgar score <5 11 0.9 33 1.1 0.88 (0.44 1.74) NS Transfer to NICU 743 67.4 2024 69.8 0.90 (0.77 1.04) NS Perinatal death 34 3.08 70 2.41 1.29 (0.83 1.99) NS Stillbirth 23 2.08 30 1.03 2.04 (1.14 3.64) 0.009 Neonatal death 11 0.99 40 1.37 0.72 (0.37 1.41) NS Assisted ventilation 85 7.7 216 7.4 1.04 (0.79 1.36) NS IC bleeding 20 1.8 68 2.3 0.77 (0.47 1.27) NS Convulsions 6 0.5 9 0.3 1.76 (0.56 5.40) NS RDS 63 5.7 155 5.3 1.07 (0.79 1.47) NS a Mann Whitney U-test NS = not statistically significant. 80
Table 5. Detailed study of 30 (1.03%) stillbirths following IVF in Flanders during the period 1997 2003 (Studiecentrum voor Perinatale Epidemiologie data). A = first born, B = second born, PPROM = preterm premature rupture of membranes, IUGR = intrauterine growth retardation, PIH = pregnancy-induced hypertension. Patient Gestational Birth Cause of fetal identification age (weeks) weight (g) death/obstetric complications 98-97-0029-A 23 640 Cervical incompetence PPROM 98-97-0029-B 23 615 Cervical incompetence PPROM 52-97-0646-A 22 510 PPROM 50-97-0268-B 37 2170 Unknown 41-97-0355-A 33 1530 PIH 18-97-0270-B 34 700 IUGR 85-98-0580-A 28 940 Solutio placentae 53-98-0563-A 23 590 Cervical incompetence PPROM 41-98-0354-A 22 580 Cervical incompetence 41-98-0354-B 22 540 Cervical incompetence 82-99-0419-B 32 1730 Unknown 40-99-0322-B 33 9519 Unknown 30-99-1957-A 23 550 PPROM 30-99-1957-B 23 750 PPROM 22-00-0648-B 23 590 Cervical incompetence 85-00-0685-B 32 660 IUGR 52-00-2221-A 31 1620 Unknown 52-00-1569-B 27 510 Unknown 41-00-0858-A 24 590 PPROM 41-00-0858-B 24 750 PPROM 85-01-1205-A 23 590 Cervical incompetence PPROM 85-01-1205-B 23 560 Cervical incompetence PPROM 76-01-0329-A 24 680 Congenital malformation 76-01-0329-B 24 540 Congenital malformation 80-02-0500-B 28 1160 Unknown 49-02-0019-B 27 930 Solutio placentae 09-02-2263-A 23 720 Cervical incompetence 85-03-0931-A 23 660 Cervical incompetence PPROM 85-03-0931-B 23 720 Cervical incompetence PPROM 52-03-1879-B 31 820 IUGR PIH 81
Table 6. Detailed study of 23 (2.08%) stillbirths following ICSI in Flanders during the period 1997 2003 (Studiecentrum voor Perinatale Epidemiologie-data). A = first born, B = second born, PPROM = preterm premature rupture of membranes, IUGR = intrauterine growth retardation, PIH = pregnancy-induced hypertension. Patient Gestational Birth Cause of fetal identification age (weeks) weight (g) death/obstetric complications 85-97-0779-A 22 570 Cervical incompetence PPROM 83-97-1133-B 32 900 IUGR 41-97-1581-A 22 500 Cervical incompetence PPROM 41-97-1581-B 22 550 Cervical incompetence PPROM 81-99-0651-B 38 2960 Unknown 43-99-0482-B 34 1560 IUGR 69-00-0060-B 36 2250 Unknown 53-00-0178-B 22 600 Cervical incompetence PPROM 40-00-0047-B 33 1860 Unknown 37-00-0354-A 35 1470 IUGR/twin-to-twin transfusion syndrome 37-00-0354-B 35 1070 IUGR/twin-to-twin transfusion syndrome 95-01-0324-A 22 660 PPROM 82-01-0944-B 34 1540 IUGR PIH 53-01-0813-A 22 610 PPROM 53-01-0813-B 22 560 PPROM 41-01-1297-A 24 610 Cervical incompetence PPROM 52-02-1637-B 26 540 PPROM 51-02-0727-B 37 1630 IUGR PIH 41-02-1444-A 23 600 Cervical incompetence PPROM 41-02-1444-B 23 550 Cervical incompetence PPROM 38-02-1096-B 37 1095 IUGR/twin-to-twin transfusion syndrome 99-03-0479-B 36 990 IUGR solutio placentae PIH 53-03-0734-A 31 570 IUGR PIH 82 exception of more prematurity and LBW cases in IVF singletons and more stillborns in the ICSI group (Tables 1 4). A study comparing the perinatal health outcome of singletons versus twins in the present study population will be part of a forthcoming paper in which ICSI pregnancies are compared with matched naturally conceived controls. Bonduelle et al. (1994, 1995), Wisanto et al. (1995, 1997) and Wennerholm et al. (1996) were the first to compare the perinatal outcome of ICSI versus IVF pregnancies. They reported similar results in IVF and ICSI pregnancies with respect to multiple pregnancy rate, low birth weight, preterm births, congenital malformations and perinatal outcome variables. In general, most studies do not show an increased risk for major birth defects in children conceived with IVF and ICSI (Van Steirteghem, 1997, 1998). More recent reports provided less reassuring results. In a world survey of ICSI pregnancies, Tarlatzis and Bili (2000) showed similar perinatal outcome including congenital malformations for IVF versus ICSI pregnancies, although the incidence of aberrations in sex chromosomes was slightly elevated for ICSI babies. This study was unfortunately a very heterogenous registry with no possibility of controlling for validity and completeness. Much more important are the results reported by Bonduelle et al. (1998, 2002). They found a statistically significant increase in sex chromosome aberrations (de-novo chromosomal anomalies) in a large series of prenatal tests in ICSI pregnancies although no increased risk for major malformations and neonatal complications was observed in the ICSI group compared with the IVF pregnancies. A two-fold increase in major birth defects with IVF and ICSI compared with naturally conceived children was reported by Hansen et al. (2002). In this study, the major birth defect rate was similar for IVF and ICSI offspring. Major malformations were found more often after multiple gestation in a large series of 5795 IVF and ICSI children reported by Bonduelle (2003). Therefore, multiple pregnancy as such seems to be responsible for the overall increase of major malformations in assisted reproduction pregnancies. According to the results of a prospective, controlled, nationwide German cohort study of more than 3000 ICSI births, the major congenital malformation rate was increased after ICSI compared with natural conception (Kataliniç et al., 2004). According to the results of Anthony et al. (2004) on congenital malformations in 4224 children conceived after IVF, the small increase in overall congenital malformations observed in the IVF children appears to be attributable to differences in maternal characteristics and not to any aspect of the IVF procedure. In a 5-year follow-up study, Place and
Englert (2003) reported a similar psychomotor and intellectual development for ICSI, IVF and spontaneously conceived children. A medical follow-up study of 5-year-old ICSI children revealed no increase in growth problems compared with children born after natural conception (Bonduelle et al., 2004). Nevertheless, concerning the question of long-term problems, some data are still missing, for example data on the fertility potential of post-pubertal children born after assisted reproduction and data on children born after intrauterine insemination and hormonal treatment (Ludwig, 2004). In the present study, the congenital malformation rate was similar between ICSI and IVF children. Compared with other reports, the incidence of congenital malformations was low (between 2.1 and 3.2%, Tables 1 4). This can partly be a result of the registration by more than 400 obstetricians and neonatologists. The data are also limited to the first 7 days after birth and it is well known that some malformations (especially minor ones) may be detected later. The manner of data collection used in the present study does not allow an exact estimation of malformations, since the registry is not standardized for this item and because minor and major malformations are not differentiated. For singleton pregnancies, similar results for all investigated perinatal variables were found. Only prematurity was seen more often in the IVF births (Tables 1 and 2). This could be the explanation for the significant difference in gestational age (P < 0.001) and birth weight (P = 0.045) between the IVF and ICSI groups. After regression analysis with correction for parity and age of the mother, the risk for LBW was significantly increased in IVF compared with ICSI pregnancies (P = 0.03). Figures for severe prematurity (less than 32 weeks) and very low birth weight did not show any difference. This means that the incidence of pregnancies at high risk for severe perinatal morbidity and subsequent longterm morbidity were not different between IVF and ICSI. Although perinatal mortality was similar in both groups, all neonatal morbidity parameters (assisted ventilation, intracranial bleeding, convulsions and respiratory distress syndrome) were seen more often in the IVF group, but the differences were not significant. A possible but hypothetical explanation may be the fact that the proportion of so-called normal women is higher in the ICSI group, since severe male subfertility is the main indication for ICSI. The proportion of unexplained infertility is probably higher in the IVF group and it is possible that this particular group is at increased risk for perinatal health problems (Pandian et al., 2001). For twin pregnancies, outcome results were similar for all investigated obstetric and perinatal parameters. The single parameter showing a difference between IVF and ICSI twins was a higher incidence of stillbirth in the ICSI group. Neonatal and perinatal mortality was similar in both groups. A detailed study of all stillbirths in IVF and ICSI pregnancies in this series showed a very high rate of extreme premature birth of less than 27 weeks due to cervical incompetence and/or preterm premature rupture of membranes (Tables 5 and 6). Cervical incompetence was seen more often in the ICSI group, and a good explanation is lacking. Intrauterine growth retardation with or without pregnancy-induced hypertension was also observed more often in the ICSI group. It was reported before that there is a higher incidence of preeclampsia in women conceiving by intrauterine insemination with washed donor spermatozoa compared with intrauterine insemination with washed partner spermatozoa. This supports, indirectly, an immunological basis for pre-eclampsia. The antigenic factor would appear to be located on the sperm as opposed to the seminal fluid itself (Smith et al., 1997; Salha et al., 1999; Wang et al., 2002). Since ICSI is mostly used in cases of azoospermia or severe oligoasthenoteratozoospermia, this could be a possible explanation of this unexpected observation. Aytoz et al. (1998) investigated the effect of sperm origin and quality on the outcome of pregnancies after ICSI. They showed that in the ejaculated sperm group, the rate of intrauterine death was higher in the severely defective sperm group than in the better-quality sperm groups. On the other hand, Ludwig and Kataliniç (2003) did not find a higher risk for pre-eclampsia in azoospermic compared with oligozoospermic patients. In the future, the database will be enlarged to investigate whether this finding is really important or not, and whether it may be linked to embryo or/and patient characteristics. Until now, there has been no information on sperm quality and its effect on obstetric outcome in the database. In conclusion, the results of this retrospective casecomparative analysis showed that the perinatal outcome of IVF and ICSI pregnancies is very similar. For singleton pregnancies, a slightly higher incidence of prematurity could be observed in the IVF group. This did not result in a significantly worse outcome in terms of perinatal morbidity and mortality. The obstetric and perinatal outcome of twin pregnancies following IVF and ICSI were also comparable with the only exception of an increased stillbirth rate in the ICSI group. A detailed study of all stillbirths in both groups showed a higher incidence of pregnancy-induced hypertension and/or intrauterine growth retardation in the ICSI group. Nevertheless, it is considered that the results of this large study are reassuring with respect to the outcome of ICSI pregnancies compared with IVF pregnancies. This conclusion holds for both singleton and twin pregnancies. Acknowledgements We gratefully acknowledge Vanessa De Bolle and Patricia Jarkowski for the technical support in preparing this manuscript. We also thank all Flemish midwifes, obstetricians and paediatricians who completed the SPEquestionnaires for many years. References Anthony S, Buitendijk SE, Dorrepaal CA et al. 2002 Congenital malformations in 4224 children conceived after IVF. Human Reproduction 17, 2089 2095. Aytoz A, Camus M, Tournaye H et al. 1998 Outcome of pregnancies after intracytoplasmic sperm injection and the effect of sperm origin and quality on this outcome. Fertility and Sterility 70, 500 505. Beral V, Doyle P, Tan SL et al. 1990 Outcome of pregnancies resulting from assisted conception. British Medical Bulletin 46, 753 768. Bergh T, Ericson A, Hillensjö T et al. 1999 Deliveries and children born after in-vitro fertilisation in Sweden 1982 95: a retrospective cohort study. Lancet 354, 1579 1585. 83
84 Blickstein I 2003 The worldwide impact of iatrogenic pregnancy. International Journal of Gynaecology and Obstetrics 82, 307 317. Bonduelle M-L 2003 ICSI-related risk for the children. PhD thesis, Free University of Brussels, pp. 41 42. Bonduelle M, Desmyttere S, Buysse A et al. 1994 Prospective follow-up study of 55 children born after subzonal insemination and intracytoplasmic sperm injection. Human Reproduction 9, 1765 1769. Bonduelle M, Legein J, Derde MP et al. 1995 Comparative followup study of 130 children born after intracytoplasmic sperm injection and 130 children born after in-vitro fertilization. Human Reproduction 10, 3327 3331. Bonduelle M, Aytoz A, Van Assche E et al. 1998 Incidence of chromosomal aberrations in children born after assisted reproduction through intracytoplasmic sperm injection. Human Reproduction 13, 781 782. Bonduelle M, Van Assche E, Joris H et al. 2002 Prenatal testing in ICSI pregnancies: incidence of chromosomal anomalies in 1586 karyotypes and relation to sperm parameters. Human Reproduction 17, 2600 2614. Bonduelle M-L, Bergh C, Niklasson A et al. 2004 Medical follow up study of 5-year-old ICSI children. Reproductive BioMedicine Online 9, 91 101. De Neubourg D, Gerris J 2003 Single embryo transfer state of the art. Reproductive BioMedicine Online 7, 615 622. Dhont M, De Sutter P, Ruyssinck G et al. 1999 Perinatal outcome of pregnancies after assisted reproduction: a case-control study. American Journal of Obstetrics and Gynecology 181, 688 695. Ericson A, Källén B 2001 Congenital malformations in infants born after IVF: a population-based study. Human Reproduction 16, 504 509. ESHRE Capri Workshop Group 2000 Multiple gestation pregnancy. Human Reproduction 15, 1856 1864. ESHRE Capri Workshop Group 2003 Mono-ovulatory cycles: a key goal in profertility programmes. Human Reproduction Update 9, 263 274. Gissler M, Silverio MM, Hemminki E 1995 In-vitro fertilization pregnancies and perinatal health in Finland 1991 1993. Human Reproduction 10, 1856 1861. Gurgan T, Demirol A 2004 Why and how should multiple pregnancies be prevented in assisted reproduction treatment programmes? Reproductive BioMedicine Online 9, 237 244. Hansen M, Kurinczuk JJ, Bower C, Webb S 2002 The risk of major birth defects after intracytoplasmic sperm injection and in vitro fertilization. New England Journal of Medicine 346, 725 730. Helmerhorst FM, Perquin DA, Donker D et al. 2004 Perinatal outcome of singletons and twins after assisted conception: a systematic review of controlled studies. British Medical Journal 328, 261 265. Jackson RA, Gibson KA, Wu YW, Croughan MS 2004 Perinatal outcomes in singletons following in vitro fertilization: a metaanalysis. Obstetrics and Gynecology 103, 551 563. Kataliniç A, Rosch C, Ludwig M 2004 Pregnancy course and outcome after intracytoplasmic sperm injection: a controlled, prospective cohort study. Fertility and Sterility 81, 1604 1616. Koivurova S, Hartikainen AL, Gissler M et al. 2002a Neonatal outcome and congenital malformations in children born after invitro fertilization. Human Reproduction 17, 1391 1398. Koivurova S, Hartikainen AL, Karinen L et al. 2002b The course of pregnancy and delivery and the use of maternal healthcare services after standard IVF in Northern Finland 1990 1995. Human Reproduction 17, 2897 2903. Koudstaal J, Braat DD, Bruinse HW et al. 2000 Obstetric outcome of singleton pregnancies after IVF: a matched control study in four Dutch university hospitals. Human Reproduction 15, 1819 1825. Lambert RD 2003 Safety issues in assisted reproductive technology: aetiology of health problems in singleton ART babies. Human Reproduction 18, 1987 1991. Lehman EL 1975 Nonparametrics: Statistical Methods Based on Ranks. Holden-Day, San Francisco. Ludwig M 2004 Development of children born after IVF and ICSI. Reproductive BioMedicine Online 9, 10 12. Ludwig M, Kataliniç A 2003 Pregnancy course and health of children born after ICSI depending on parameters of male factor infertility. Human Reproduction 18, 351 357. Luke B, Keith LG 1992 The contribution of singletons, twins and triplets to low birth weight, infant mortality and handicap in the United States. Journal of Reproductive Medicine 37, 661 666. McGovern PG, Llorens AJ, Skurnick JH et al. 2004 Increased risk of preterm birth in singleton pregnancies resulting from in vitro fertilization-embryo transfer or gamete intrafallopian transfer: a meta-analysis. Fertility and Sterility 82, 1514 1520. National Perinatal Statistics Unit and Fertility Society of Australia 1988 IVF and GIFT pregnancies: Australia and New Zealand. Sydney National Perinatal Statistics Unit. Ombelet W, De Sutter P, Van der Elst J, Martens G 2005 Multiple gestation and infertility treatment: registration, reflection and reaction: the Belgian project. Human Reproduction Update 11, 31 34. Palermo G, Joris H, Devroey P, Van Steirteghem AC 1992 Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet 340, 17 18. Pandian Z, Bhattacharya S, Templeton A 2001 Review of unexplained infertility and obstetric outcome: a 10 year review. Human Reproduction 16, 2593 2597. Patrizio P 1995 Intracytoplasmic sperm injection (ICSI): potential genetic concerns. Human Reproduction 10, 2500 2523. Place I, Englert Y 2003 A prospective longitudinal study of the physical, psychomotor, and intellectual development of singleton children up to 5 years who were conceived by intracytoplasmic sperm injection compared with children conceived spontaneously and by in vitro fertilization. Fertility and Sterility 80, 1388 1397. Reubinoff BE, Samueloff A, Ben-Haim M et al. 1997 Is the obstetric outcome of in vitro fertilized singleton gestations different from natural ones? A controlled study. Fertility and Sterility 67, 1077 1083. Salha O, Sharma V, Dada T et al. 1999 The influence of donated gametes on the incidence of hypertensive disorders of pregnancy. Human Reproduction 14, 2268 2273. Schieve LA, Meikle SF, Ferre C et al. 2002 Low and very low birth weight in infants conceived with use of assisted reproductive technology. New England Journal of Medicine 346, 731 737. Smith GN, Walker M, Tessier JL, Millar KG 1997 Increased incidence of preeclampsia in women conceiving by intrauterine insemination with donor versus partner sperm for treatment of primary infertility. American Journal of Obstetrics and Gynecology 177, 455 458. Strömberg B, Dahlquist G, Ericson A et al. 2002 Neurological sequelae in children born after in-vitro fertilisation: a populationbased study. Lancet 359, 461 465. Tan SL, Doyle P, Campbell S et al. 1992 Obstetric outcome of in vitro fertilization pregnancies compared with normally conceived pregnancies. American Journal of Obstetrics and Gynecology 167, 778 784. Tanbo T, Dale PO, Lunde O et al. 1995 Obstetric outcome in singleton pregnancies after assisted reproduction. Obstetrics and Gynecology 86, 188 192. Tarlatzis BC, Bili H 2000 Intracytoplasmic sperm injection. Survey of world results. Annals of the New York Academy of Sciences 900, 336 344. Templeton A 2003 IVF success rates the importance of data presentation. Reproductive BioMedicine Online 7, 524 526. Van Steirteghem A 1997 Intracytoplasmic sperm injection. Baillieres Clinics in Obstetrics and Gynaecology 11, 725 738. Van Steirteghem A 1998 Outcome of assisted reproductive technology. New England Journal of Medicine 338, 194 195. Verlaenen H, Cammu H, Derde MP, Amy JJ 1995 Singleton pregnancy after in vitro fertilization: expectations and outcome. Obstetrics and Gynecology 86, 906 910. Wang JX, Knottnerus AM, Schuit G et al. 2002 Surgically obtained sperm, and risk of gestational hypertension and pre-eclampsia. Lancet 359, 673 674.
Wennerholm UB 2003 Health and development of twins, triplets and higher-order multiple births. Reproductive BioMedicine Online 7, 518 520. Wennerholm UB, Bergh C, Hamberger L et al. 1996 Obstetric and perinatal outcome of pregnancies following intracytoplasmic sperm injection. Human Reproduction 11, 1113 1119. Wennerholm UB, Bergh C, Hamberger L et al. 2000 Incidence of congenital malformations in children born after ICSI. Human Reproduction 15, 944 948. Wisanto A, Magnus M, Bonduelle M et al. 1995 Obstetric outcome of 424 pregnancies after intracytoplasmic sperm injection. Human Reproduction 10, 2713 2718. Wisanto A, Bonduelle M, Camus M et al. 1996 Obstetric outcome of 904 pregnancies after intracytoplasmic sperm injection. Human Reproduction 11 (Suppl. 4), 121 129. Received 10 January 2005; refereed 25 January 2005; accepted 26 April 2005. 85