Portal de Periódicos da CAPES

High frequency of iatrogenic monozygotic twins with administration of clomiphene citrate and a change in chorionicity

2006; Elsevier BV; Volume: 85; Issue: 3 Linguagem: Inglês

10.1016/j.fertnstert.2005.08.034

1556-5653

Cathérine Derom, Fernand Leroy, Robert Vlietinck, Jean‐Pierre Fryns, R. Derom,

Ovarian function and disorders

The data of the population-based, prospective survey of multiple births in East Flanders, Belgium, show that a greater proportion of monozygotic twins is observed with clomiphene citrate as sole treatment compared with other ovulatory drugs. Among iatrogenic monozygotic twins, the ratio of monochorionic versus dichorionic pairs is higher compared with spontaneous twins. The data of the population-based, prospective survey of multiple births in East Flanders, Belgium, show that a greater proportion of monozygotic twins is observed with clomiphene citrate as sole treatment compared with other ovulatory drugs. Among iatrogenic monozygotic twins, the ratio of monochorionic versus dichorionic pairs is higher compared with spontaneous twins. Because of the well-known difficulties involved in obtaining a single ovulation in the treatment of anovulation and the practice of multiple embryo replacement in assisted reproductive technologies (ART), modern methods of infertility treatment result in marked modifications of the expected number of twin and triplet births (1Derom C. Maes H. Derom R. Van den Berghe H. Vlietinck R. Iatrogenic multiple pregnancies in East Flanders, Belgium.Fertil Steril. 1993; 60: 493-496PubMed Google Scholar, 2Jones H.W. Multiple births how are we doing?.Fertil Steril. 2003; 79: 17-21Abstract Full Text Full Text PDF PubMed Scopus (55) Google Scholar). In the last two or three decades, a progressive increase in the frequency of twins has occurred, whereby the total number of twin births has almost doubled in East Flanders, Belgium. The rise in the number of twin births is not unexpected as regards the rate of dizygotic (DZ) twins, but the simultaneous rise of monozygotic (MZ) twinning is a surprise, and a satisfactory explanation has yet to be found. Among iatrogenic cases, the true frequency of MZ twinning is obscured by multiple embryo transfer and early embryonic demise, the impact of the latter being difficult to quantitate. A great problem related to the study of MZ twinning in relation to infertility treatment has been the incomplete determination of the different MZ chorionicity types, both from individual infertility clinics (3Schieve L.A. Meikle S.F. Peterson H.B. Jeng G. Burnett N.M. Wilcox L.S. Does assisted hatching pose a risk for monozygotic twinning in pregnancies conceived through in vitro fertilization?.Fertil Steril. 2000; 74: 288-294Abstract Full Text Full Text PDF PubMed Scopus (149) Google Scholar, 4Sills E.S. Tucker M.J. Palermo G.D. Assisted reproductive technologies and monozygous twins implications for future study and clinical practice.Twin Res. 2000; 3: 217-223Crossref PubMed Scopus (56) Google Scholar) or from pooled data (5Edwards R.G. Mettler L. Walters D.E. Identical twins and in vitro fertilization.In Vitro Fert Emb Transf. 1986; 3: 114-117Crossref PubMed Scopus (156) Google Scholar, 6Slotnick R.N. Ortega J.E. Monoamniotic twinning and zona manipulation a survey of US IVF centers correlating manipulation procedures and high-risk twinning frequency.J Assist Reprod Genet. 1996; 13: 381-385Crossref PubMed Scopus (68) Google Scholar). At best, monochorionic (MC) and the rare monoamniotic (MA) twins or higher multiples were identified echographically and/or morphologically at birth, thereby overlooking dichorionic (DC) MZ twins within the large numbers of same-sex DZ twin pairs. In general populations, about one third of spontaneous MZ twins are dichorionic (7Bulmer M.C. The biology of twinning. Oxford University Press, London1970Google Scholar, 8Machin G.A. Placentation in multiple births.Twin Res. 2002; 4: 150-155Google Scholar). The present study aims at a better evaluation of the frequency and placentation of MZ twinning in relation to ART and to the use of various ovulatory drugs. The population-based data collected since 1964 by the East Flanders Prospective Twin Survey (EFPTS) provides accurate zygosity and chorionicity determinations for almost all pairs of twins and triplets born in the province (9Loos R. Derom C. Vlietinck R. Derom R. The East Flanders Prospective Twin Survey (Belgium) a population-based register.Twin Res. 1998; 1: 167-175PubMed Google Scholar). Furthermore, it distinguishes between spontaneous cases and those resulting from ART or the use of ovulatory drugs alone (non-ART). Between 1964 and 2002, the register had enrolled 6317 twin pairs and 171 sets of triplets. For the present analysis, twin pairs that resulted from selective embryo-reduction (n = 60), and, since 1985, twin (n = 49) and triplet maternities (n = 1) of unknown origin were excluded. Eventually, 6208 twin pairs and 170 triplet sets were analyzed. Of these, 1528 twin and 132 triplet maternities were induced. All dichorionic MZ twins and trichorionic DZ triplets showed a probability of monozygosity of at least 98%. The contingency chi-square test was used for comparison of data. Analyses were conducted with the SAS 6.12 computer package (SAS Institute, Inc., Cary, NC). Reported P values are two-sided; P=<.05 was considered statistically significant. Table 1 illustrates the distribution of zygosity types among spontaneous and induced twin and triplet maternities registered between 1964 and 2002. A marked difference appears in that the relative proportions of dizygotic and trizygotic cases were statistically significantly higher among induced twin and triplet maternities, respectively (P<.001). The use of ovulatory drugs alone induces a relative proportion of MZ twins, which is statistically significantly higher than that observed after ART (7% vs. 2%) (P<.001). Although showing the same trend, the difference in this regard between DZ triplets was not statistically significant (14% vs. 9%).TABLE 1Zygosity of spontaneous and iatrogenic twin and triplet maternities (1964–2002).ZygosityTwinsTripletsSpontaneousAIO onlyARTSpontaneousAIO onlyARTTrizygotic———9 (24%)74 (86%)38 (83%)Dizygotic2529 (54%)704 (92%)738 (97%)23 (60%)12 (14%)4 (9%)Monozygotic2072 (44%)57 (7%)17 (2%)6 (16%)0 (0%)1 (2%)Unknown79 (2%)4 (1%)8 (1%)0 (0%)0 (0%)3 (6%)Total4680 (100%)765 (100%)763 (100%)38 (100%)86 (100%)46 (100%)Note: Difference between spontaneous and iatrogenic: P<.001 for twins and triplets. Difference between AIO only and ART: P<.001 for twins; non significant for triplets. AIO only = artificial induction of ovulation solely; ART = assisted reproduction techniques.Derom. Etiology and placentation in iatrogenic MZ twins. Fertil Steril 2006. Open table in a new tab Note: Difference between spontaneous and iatrogenic: P<.001 for twins and triplets. Difference between AIO only and ART: P<.001 for twins; non significant for triplets. AIO only = artificial induction of ovulation solely; ART = assisted reproduction techniques. Derom. Etiology and placentation in iatrogenic MZ twins. Fertil Steril 2006. Among all MZ twins, the frequency of MC placentas in all induced cases (ART and non-ART) was higher than in spontaneous cases (80% versus 64%; P=.02). In the non-ART group, clomiphene citrate (CC) given alone was the most frequent treatment associated with twins (43%), and gonadotrophins without additional treatment are second in frequency (28%). Among ART cases, gonadotrophins added to a gonadotrophin-releasing hormone (GnRH) analog or pulsatile GnRH is the most frequent therapy related to twinning cases (34%); gonadotrophins alone accounted for 21% of cases in this group. Clomiphene citrate given alone represented less than 1% of the cases. We further focused on the association of pure CC treatment in relation to zygosity of twins and triplets. As almost no such cases (less than 1%) occurred in the ART group, this latter was excluded from the analysis. Cases where the only treatment was CC (324 twins and 23 triplets) were compared with iatrogenic cases where no CC (304 twins and 46 triplets) was used. Cases that resulted from a combination of gonadotrophins and CC were excluded: 137 twin and 17 triplet maternities. A statistically significant greater proportion of MZ twins was observed with CC as sole treatment as compared with other cases in which no CC was used (12% vs. 3.6%) (P<.001). Although showing the same trend, the difference in this regard between DZ triplets was not statistically significant (22% vs. 9%). The relative differences in MZ/DZ proportions associated with different modes of infertility treatment in this study can mostly be explained by the variability of DZ twinning rates associated with these different treatments: around 25% after ART, 17% after gonadotrophins, 8% after CC, and 0.7% in spontaneous cases. In fact, the MZ/DZ proportion is expected to be highest in spontaneous cases and higher in cases after treatment with CC (compared with cases after treatment with gonadotrophins) and lowest after ART. It should be noted that in this study we focus on the relative difference in MZ and DZ twin frequencies and not on MZ twinning rates, terms that are often confounded. The MZ twinning rate is to be calculated by dividing the number of MZ births by the number of births from the same mode of conception. In this context, Blickstein et al. (10Blickstein I. Jones M. Keith L.G. Zygotic-splitting rates after single-embryo-transfers in in vitro fertilization.N Engl J Med. 2003; 348: 2366-2367Crossref PubMed Scopus (80) Google Scholar) rightly emphasized that the evaluation of twinning among maternities occurring after transfer of a single embryo is one of the safest methods for determining the rate of monozygosity related to ART. The investigators reported a 2.3% rate of zygotic splitting among in vitro conceptions after single-embryo transfer, a figure that is six times the rate of 0.4% quoted in the literature for spontaneous conceptions (7Bulmer M.C. The biology of twinning. Oxford University Press, London1970Google Scholar). As far as the type of drug is concerned, our data show that in the category of simple ovulation induction or enhancement, CC, which has been the most widely used substance, induced a significantly higher MZ to DZ proportion than observed in other treatment groups. As already mentioned, this result is obviously related to the mono-ovulatory aim of this treatment and to its relatively lower power to induce multiple follicle maturation and DZ twinning (around 8%) (11Adashi E. Clomiphene the coming of a monoisomeric preparation.in: Evers J.L. Ovulation induction the difficult patient. Baillière Tindall, London1993: 331-348Google Scholar). However, the possibility remains that such difference might be partly related to an intrinsic capacity of CC to induce monozygosity. Clomiphene citrate is structurally related to diethylstilbestrol. As a long-acting drug with a half-life of 5 days, it is said to remain active in the cycle following the one during which it was administered (12Geier A. Lunenfeld B. Pariente C. Kotev-Emeth S. Shadani A. Kokaia E. Estrogen receptor binding material in blood of patients after clomiphene citrate administration determination by a radioreceptor assay.Fertil Steril. 1987; 47: 778-784Abstract Full Text PDF PubMed Google Scholar). From the beginning of its use in the 1950s, this compound has been suspected of exerting teratogenic effects, but related epidemiologic investigations have remained inconclusive so far (13Sorensen H.T. Pedersen L. Skriver M. Norgaard M. Norgard B. Hatch E. Use of clomiphene during early pregnancy and risk of hypospadias population-based case-control study.BMJ. 2004; 330: 126-127Crossref PubMed Scopus (36) Google Scholar). Some experimental data, however, have pointed at a possible cytotoxic effect on the developing zygote (11Adashi E. Clomiphene the coming of a monoisomeric preparation.in: Evers J.L. Ovulation induction the difficult patient. Baillière Tindall, London1993: 331-348Google Scholar, 14Birkenfeld A. Beier H. Schenker J.G. The effect of clomiphene citrate on early embryonic development, endometrium and implantation.Hum Reprod. 1986; 1: 387-395PubMed Google Scholar). The possible impact of CC on MZ twinning should be further investigated. Obstetricians should keep in mind that, cases of a twin pregnancy after CC treatment have about one chance out of 7 or 8 to result in MZ twins. Furthermore, among MZ twins the ratio of MC (late embryo splitting) versus DC pairs (early embryo splitting) is higher in the iatrogenic (ART and non-ART) group in general. This surely needs further investigation and attention because no changes in the ratio had been observed previously. The increased risk of embryonic splitting associated with infertility treatment is not merely of academic interest—the outcome of twin maternities are markedly affected by zygosity, and more specifically chorionicity. Several disease processes are found in MC twins that do not occur in DC twins. Also, MC twins are notoriously prone to suffer from a number of special complications such as malformations, twin-to-twin transfusion with deleterious hemodynamic imbalance, or fetal growth discordance and restriction (8Machin G.A. Placentation in multiple births.Twin Res. 2002; 4: 150-155Google Scholar, 15Leroy F. Les jumeaux dans tous leurs états. Brussels: Ed. Deboeck Université, 1995Google Scholar). The risk for perinatal mortality and morbidity is 3 to 5 times higher than for a dichorionic twin pregnancy (9Loos R. Derom C. Vlietinck R. Derom R. The East Flanders Prospective Twin Survey (Belgium) a population-based register.Twin Res. 1998; 1: 167-175PubMed Google Scholar). In addition, MC twinning is problematic when considering fetal reduction. Thus, the increased risk of MC twinning needs special obstetrical attention. In a recent review of 81 MZ twin pregnancies after in vitro fertilization/embryo transfer (IVF-ET), Alikani et al. (16Alikani M. Cekleniak N.A. Walters E. Cohen J. Monozygotic twinning following assisted conception an analysis of 81 consecutive cases.Hum Reprod. 2003; 18: 1937-1943Crossref PubMed Scopus (128) Google Scholar) also reported an increased risk of MC twinning after ART, suggesting that the underlying causes of MZ twinning after ART manifest themselves relatively late in embryogenesis. This could be an explanation for the high MC/DC ratio found in our series of iatrogenic MZ twins. They reported an unexpected increase in the rate of monoamniotic twinning, a rare event that only happens in 2% to 3% of all spontaneous MZ twin pregnancies. We could not confirm their data; in our study of 74 MZ twins, 16 DZ triplets, and 1 MZ triplet, we had only two cases of monoamniotic twinning (one DZ triplet after ART and one MZ twin after CC). It should be noted that in our series placentation is determined at birth and that only twin pairs, from whom at least one of the children weighed 500 g or more or gestated at least 22 weeks, were considered; in their series, MZ twin pregnancies are identified by ultrasound examination at around 7 weeks' gestation. Other explanations could be the high antenatal mortality rate among these twins or the possibility of false-positive diagnosis of monoamnionicity with ultrasound diagnosis (17Strohbehn K. Dattel B.J. Pitfalls in the diagnosis of nonconjoined monoamniotic twins.J Perinat. 1995; 15: 484-493PubMed Google Scholar). The authors thank Jana Vandenbergh and Gert Matthijs for excellent technical assistance.