Monozygotic twinning after assisted reproductive techniques: a phenomenon independent of micromanipulation

Morey Schachter1, Arieh Raziel, Shevach Friedler, Devorah Strassburger, Orna Bern and Raphael Ron-El

IVF and Infertility Unit, Assaf Harofeh Medical Center, Zerifin, The Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A 3 year retrospective analysis was conducted of pregnancies achieved after various assisted reproductive treatment modalities in our infertility practice, to calculate and compare the rates of monozygotic twinning (MZT). A total of 731 pregnancies achieved after various assisted reproduction treatments were reviewed. Gonadotrophin therapy for induction of ovulation and controlled ovarian hyperstimulation (COH) yielded 129 clinical pregnancies. Conventional IVF yielded 139 pregnancies. IVF and intracytoplasmic sperm injection (ICSI) with or without assisted hatching (AH) yielded 463 pregnancies, all during the same time period. The rates of multiple pregnancy (monozygotic and dizygotic) twins and triplets were recorded. MZT was found in 1.5% of ovulation induction or COH pregnancies (2/129). The incidence of MZT after conventional IVF was 0.72% (1/139). After IVF–ICSI/AH, MZT was found in 0.86% (4/463). The overall rate of MZT was 0.95% (7/731). Five cases were dizygotic triplets and two cases were monozygotic twins. We found the rate of MZT after assisted reproduction treatment increased more than two-fold over the background rate in the general population. Dizygotic triplets were found more often than monozygotic twins. The rate of MZT was consistently increased, irrespective of treatment modality or micromanipulation. This may signify that the aetiology of increased MZT after assisted reproduction is the gonadotrophin treatment rather than in-vitro conditions, micromanipulation, or multiple embryo transfer.

Key words: gonadotrophin therapy/micromanipulation/monozygotic twins/multiple pregnancy/zona pellucida


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Factors predisposing to the production of dizygotic twins are known to include maternal age, parity, race, family history of dizygotic twinning and of course treatment with drugs that induce multiple ovulation and related assisted reproductive techniques. In contrast, the factors that affect the frequency of monozygotic twinning (MZT) have been poorly characterized. Previous reports have linked the use of ovulation induction drugs to MZT (Derom et al., 1987Go). Other reports (Edwards et al., 1986Go; Alikani et al., 1994Go; Slotnick and Ortega, 1996Go; Hershlag et al., 1999; Blickstein et al., 1999Go; Sills et al., 2000Go) have calculated an increased frequency of MZT in the setting of IVF, with or without zona manipulation. One hypothesis advanced in order to explain the higher incidence of monozygosity suggested that manipulation of the zona pellucida could encourage inner cell mass herniation during hatching. Another hypothesis (Sills et al., 2000Go) discussed the increased incidence of twinning as simply a function of the presence of more embryos in the uterine cavity after embryo transfer.

We report our experience with monozygotic twinning after ovulation induction or assisted reproduction treatment, from different clinical settings, and advance our hypothesis as to the evolution of this situation.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A total of 731 pregnancies achieved after employing various assisted reproduction procedures over a 3 year period (1997–1999) at the Assaf Harofeh Medical Center IVF and Infertility Unit were reviewed. Induction of ovulation and controlled ovarian stimulation with gonadotrophins were administered in 480 cycles to 220 patients in this time period. The mean age of these patients was 30.2 ± 6.7 years. Tubal patency (at least one open tube) was demonstrated in all patients by hysterosalpingography prior to initiating gonadotrophins. Semen analysis carried out on all patients' partners were within normal limits (85%) or suboptimal (15%). The indications for treatment included anovulatory infertility in 72% (158 patients) and unexplained infertility in 28% (62 patients). Gonadotrophin treatment was initiated after an initial unsuccessful trial utilizing clomiphene citrate in doses ranging from 50–200 mg/day for 5 days per cycle in 2–6 previous cycles per patient. The gonadotrophins used were human menopausal gonadotrophin (HMG, Pergonal®; Teva, Petah Tikva, Israel) in 365 cycles (76%) and purified urinary follicle stimulating hormone (pFSH, Metrodin®; Teva) in 115 cycles (24%). Doses ranged from 75 IU/day to no more than 225 IU/day for the duration of treatment ranging from 5–18 days per cycle (median 8 days). The final stage of ovulation was induced using human chorionic gonadotrophin (HCG, Chorigon®; Teva) in all cycles, when ovarian follicles with a diameter of >=19 mm were identified by transvaginal ultrasound. The dose of HCG varied from 5000–10000 IU injected i.m. Serum ßHCG was assessed in all patients who did not menstruate by day 16 after HCG administration. At 1, 2 and 3 weeks after a positive pregnancy test was returned, transvaginal ultrasound was carried out, and the number of gestational sacs and embryos recorded. Monozygosity can only be proven in the case of a single embryo transferred, so in effect it is the chorionicity of the pregnancy that can be demonstrated by ultrasound. However, as all previous studies did not make this distinction, for the purpose of uniformity and clarity, we used the term monozygosity where chorionicity would have been more absolutely accurate. When more than one sac or embryo was visualized, the chorionicity/zygosity of the embryos was assessed, using the criteria of Townsend et al. (Townsend et al., 1988Go) and multiple pregnancies were followed in our unit until the 9th gestational week. In early pregnancy, the determination of chorionicity by ultrasound is based mainly on the inter-twin membrane thickness, as dichorionic septae have a mean thickness of 2.4 mm as opposed to 1.4 mm in monochorionic septae (Winn et al., 1989Go), and a 100% accuracy rate was reported by Monteagudo et al. using these criteria at 9 gestational weeks (Townsend et al., 1988Go; Monteagudo et al., 1994Go).

During the same time period, 602 pregnancies were achieved in our IVF programme. A total of 475 cycles of conventional IVF yielded 139 clinical pregnancies (29.3%). Conventional IVF was employed in patients with tubal factor infertility or unexplained infertility after gonadotrophin therapy with intrauterine inseminations had failed. Tubal factor infertility was determined by pathological hysterosalpingography or previous laparoscopy/laparotomy, and was so determined in 211 of these 299 patients, including 39 patients with endometriosis. Ninety-five patients were diagnosed as suffering from unexplained infertility. IVF was carried out after pituitary down-regulation with gonadotrophin-releasing hormone agonists (GnRHa, Decapeptyl® 3.75 controlled release; Ferring, Malmo, Sweden), or nasal Nafarelin® 400–600 µg/day (Synarel, Delpharm, France) and HMG, using a standardized long protocol. The administration of HCG, oocyte aspiration, laboratory procedures, semen preparation, embryo handling and transfer have been described in detail (Ron-el et al., 1991Go). Embryo transfer was achieved in 92.4% of attempted cycles, the average number of embryos replaced was 2.4 ± 0.8 embryos per cycle. Identification of pregnancy and subsequent ultrasound examinations were as for the ovulation induction group.

Micromanipulation was employed in our IVF programme in 1438 cycles in the same time period. Intracytoplasmic sperm injection (ICSI) was used for patients with male factor infertility—defined as severe oligoasthenoteratozoospermia—or in cases of failed fertilization in a previous conventional IVF cycle with a normal ovarian response. ICSI was also employed in cases of poor response on the part of the female partner (four oocytes or less aspirated per cycle) when two (or more) previous IVF attempts had been unsuccessful. Assisted hatching (AH) using acid Tyrode's solution (Irvine Scientific, Santa Anna, CA, USA) was employed in 327 of the 1438 ICSI cycles (22.7%). All of these treatment cycles were initiated using the same standard drug protocols, criteria for HCG administration, oocyte aspiration, laboratory procedures and semen preparation as previously described. The ICSI procedure was carried out as described by Van Steirteghem et al. (Van Steirteghem et al., 1993Go). Embryo transfer was achieved in 94% of attempted cycles, the average number of embryos replaced was 2.3 ± 1.2. Identification of pregnancy and subsequent ultrasound examination were as for the previously mentioned groups. Micromanipulation (ICSI and/or AH) in 1438 embryo transfer cycles yielded 463 clinical pregnancies (32.2%), the proportion of AH in conception and non-conception cycles was similar.

Statistical analyses
Variables relating to the patient or treatment characteristics were examined and compared with the overall IVF patient population treated in our centre, using Fisher's exact test to compare rates (drug type and dose distribution) and Mann–Whitney Rank Sum test, to compare non-parametric data (age, peak oestradiol and oocytes aspirated).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
A total of 731 pregnancies was achieved in the three assisted reproduction arms, of these, 126 clinical multiple pregnancies were identified (Table IGo), giving an overall multiple pregnancy rate of 17.2%.


View this table:
[in this window]
[in a new window]
 
Table I. Distribution and types of multiple pregnancies in 731 pregnancies after assisted reproduction treatment
 
In the ovulation induction/COS group, the multiple pregnancy group comprised 10.8% (14/129), of these, twins comprised ten sets of dizygotic twins and one set of monozygotic twins, triplets comprised two sets of trizygotic and one set of dizygotic triplets. Therefore, the overall rate of MZT in this group was 1.5% (2/129).

The multiple pregnancy rate in the conventional IVF arm was found to be 21.5% (30/139). Twins comprised 26 sets of 30 multiple pregnancies, all of which were dizygotic. Four sets of triplets were found in this group—three were trizygotic and one dizygotic. The overall rate of MZT in this group was 0.72% (1/139).

The addition of micromanipulation (ICSI with or without AH) to IVF resulted in 463 pregnancies. Of these, 82 were multiple pregnancies (17.7%). Twins comprised 63 of the 82 multiple pregnancies, 62 were dizygotic and one was monozygotic. Triplets were diagnosed in 19 of these 463 pregnancies, 16 were trizygotic and three dizygotic. The overall rate of monozygotic twinning in the IVF–micromanipulation group was 0.86% (4/463). If the rate of monozygosity is calculated by dividing the number of monozygotic twins found by the number of implanted embryos (and not the number of pregnant patients), the `true' rate of monozygosity is slightly lower, as the denominator takes into account the dizygotic twins as two implantations, although this calculation was not made in previous studies (Table IGo).

Nearly 80% of the 126 sets of twins and triplets reached live delivery (99/126). Precise data on all placentae and membranes were not available.

Spontaneous reduction to singleton from twins or to twins from triplets occurred in five cases. Interventional reduction of triplets to twins in two cases of trizygotic triplets led to one late abortion of all three triplets, and to one ongoing twin pregnancy, which delivered at term. Analysis of the seven cases of MZT did not disclose any specific pattern, either in patient or in treatment variables, that could be construed as indicative of a monozygotic-prone profile. Patients' ages and indication for treatment were representative of the whole group, as were the type of medication used, dose of HCG, peak oestradiol levels, number of oocytes aspirated or number of embryos replaced. Monozygotic patients' demographic, treatment and outcome data are concluded in Table IIGo.


View this table:
[in this window]
[in a new window]
 
Table II. Patients' clinical treatment and outcome data
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The mechanism of MZT is as yet poorly understood. In the simplest terms, monozygotic twins arise from division of the fertilized ovum at various early stages of development of the embryo. When division of the embryonic cell mass occurs earlier than 72 h after fertilization, biamniotic bichorionic monozygotic twins will evolve. Division of the embryo after the inner cell mass has formed, between day 4 and day 8, will give rise to biamniotic monochorionic twins. Splitting after day 8 will lead to mono-amniotic monochorionic twins. This process probably begins with the protrusion of some tropho-ectoderm cells through a small opening in the zona pellucida (ZP). Some cells of the inner cell mass may then break off in utero to form monozygous twins (Malter and Cohen, 1989Go). Multiple gaps in the ZP may even lead to multiple herniation, possibly contributing to higher-order monozygotic pregnancies (Cohen and Feldberg, 1991Go).

Six of our cases showed a biamniotic monochorionic membrane configuration for the monozygotic twins and the seventh case was demonstrated to have a mono-amniotic–monochorionic membrane configuration for the twins, which demonstrates the time-frame of embryo splitting during hatching. Two of the seven cases demonstrated mono-amniotic twins in a dizygotic triplet (cases 2 and 3, Table IIGo). This is rare event, and both patients had an extremely premature delivery.

Previous reports have described an increased incidence of MZT with assisted reproduction techniques (Table IIIGo). The incidence of MZT in the general population has been calculated to be ~1/250 (0.42%) live births, and could actually be higher if all clinical pregnancies were included (Bulmer, 1970Go; Saito et al., 2000Go).


View this table:
[in this window]
[in a new window]
 
Table III. Monozygosity after assisted reproduction – previous publications
 
A frequency of 1.2% for MZT after treatment with ovulation induction drugs (clomiphene citrate and gonadotrophins) was observed by Derom and colleagues, which was significantly higher than the expected frequency of 0.45% after spontaneous ovulation (Derom et al., 1987Go, 1993Go). Edwards and colleagues calculated the incidence of MZT in the population of IVF patients to be 1.33% (eight MZT in 600 conventional IVF pregnancies worldwide in 1986, the calculated 95% confidence interval 0.58–2.63%) (Edwards et al., 1986Go). The authors concluded that the artificial conditions of in-vitro media are the likely causes of increased incidence of MZT in this population. Recently, Blickstein and colleagues (1999) reported their experience in the setting of conventional IVF, in the special case of single embryo transfer, enabling an absolutely reliable diagnosis of MZT. They found the incidence of MZT in conventional IVF to be 5% (4/82) whereas no cases of MZT were found after ICSI (0/94) (Blickstein et al., 1999Go).

Alikani et al. (1994) reported their experience with MZT in 737 pregnancies achieved after IVF–embryo transfer, ~75% of these arose after various forms of zonal manipulation (Alikani et al., 1994Go). These included subzonal insemination (SUZI), partial zona dissection (PZD), zonal drilling, or AH of embryos. Six sets of MZT or triplets were described, resulting in an incidence of 0.84%, approximately twice the expected frequency. Here, the authors propound zonal manipulation procedures as a possible cause of MZT.

Zonal manipulation was also cited (Slotnick and Ortega, 1996Go) as a major factor in the increased incidence of mono-amniotic twins in the IVF–embryo transfer (micromanipulation) population. They found an incidence of 3.49% (5 /143), more than seven times the background rate of twinning. Hershlag and colleagues (1999) noted a significantly increased rate of MZT after mechanical assisted hatching after conventional IVF, which reached a rate of 1.2% per embryo transfer (Hershlag et al., 1999). Schieve et al. (2000) calculated the odds ratio of monozygotic twinning after assisted hatching in assisted reproduction pregnancies to be 3.2–3.8, and they considered the contribution of AH to be significantly beyond that of the ovulation induction drugs (Schieve et al., 2000Go).

Wenstrom et al. (1993) counted seven monozygotic pregnancies of a total of 218 (3.2%). (Wenstrom et al., 1993Go) These constituted 9.8% of all multiple gestations, whereby the type of assisted reproduction treatment had no effect on the incidence of monozygotic twinning. The authors theorized that assisted reproduction has an effect on the timing of embryonic events that lead to an increased incidence of MZT, regardless of the specific assisted reproduction used.

Additional reports have highlighted the possibility of higher-order monozygous pregnancies after assisted reproduction treatment. Three cases of dizygotic triplet pregnancy, in addition to those reported here, have been described in detail (Avrech et al., 1993Go; Steiner and Ojakangas, 1994Go; Inion et al., 1998Go). All three arose after IVF without micromanipulation. Avrech et al. (1993) reported good obstetric results, all three babies delivered thrived, whereas Steiner and Ojakangas (1994) reported delivery at 28 weeks with subsequent neonatal death of two of the three infants. Inion et al. (1998) described monozygotic intrauterine twins and a tubal pregnancy after replacement of two embryos. This combination of heterotopic pregnancy and dizygotic triplets is extremely rare. A trizygotic quadruplet pregnancy has also been described (Biljan et al., 1995Go) after transfer of three grade I embryos after IVF, again without micromanipulation. Salat-Baroux and colleagues (1994) described trizygotic quintuplets (mono-amniotic triplets with two additional sacs and embryos) after transfer of four grade I embryos after IVF without micromanipulation (Salat-Baroux et al., 1994Go). This patient aborted after an attempt at fetal reduction at 13 gestational weeks. These cases demonstrate that higher-order MZT may also be independent of micromanipulation.

Multicentric experience with MZT after blastocyst transfer has been reported (Behr et al., 1999Go). The incidence of MZT in this specific setting was calculated to be 5% (10/199), which is higher than the background incidence of MZT by a factor of 10. Abusheikha et al. (2000) reported on 11 monozygotic multiple pregnancies from two assisted reproduction centres of 718 pregnancies (Abusheikha et al., 2000Go). Six of the eleven occurred after ICSI.

Finally, Sills and colleagues reported on 23 sets of monozygotic twins from 1911 assisted reproduction pregnancies (1.2%) (Sills et al., 2000Go). Their data confirm that no correlation was found between zonal manipulative techniques and MZT. This group of investigators concluded that the most likely cause of MZT in the setting of assisted reproduction was the increased number of embryos transferred to the uterus. The rate of MZT was approximately three times the background rate, closely reflecting the number of embryos transferred per cycle.

A number of hypotheses may be advanced in an effort to explain the increased incidence of MZT in micromanipulation or blastocyst cycles. Alikani et al. (1994) suggested that artificially induced structural changes in the ZP may lead to complications in the normal process of zonal lysis (Alikani et al., 1994Go). The embryo might bypass its own mechanism of local zonal lysis and `choose' to escape through the already established path. The dimensions of this gap could restrict the emerging embryo and encourage splitting, leading to MZT. A naturally thin or irregular zona pellucida may affect the hatching process in the same way, if during the blastocyst stage the zona disintegrates at more than one spot. This could be one of the mechanisms to explain the age-related increase of MZT, as zona thickness is more irregular with age (Cohen and Feldberg, 1991Go).

Edwards and colleagues discussed the incidence of MZT after IVF without zonal manipulation (Edwards et al., 1986Go). Importantly, they noted that most MZT arise spontaneously in vivo from a single embryo (monozygotic twins), conversely, many MZT arising from IVF were accompanied by one or more sibling embryos, i.e., monozygotic triplets or higher order twinning. They concluded that the nature of embryonic growth in vitro predisposes to twinning, as opposed to in-vivo growth after ovarian stimulation. A possible `hardening' of the human ZP in vitro, after exposure to artificial media, as opposed to salpingeal or uterine secretions, could lead to increased fragility or brittleness and thence `fracture' the ZP, or cell-to-cell adhesion might be disturbed after in-vitro culture. Sills et al. (2000) postulated that the increased rate of MZT is a simple reflection of an increase in the number of embryos transferred to the uterus, and that the rate of MZT per transferred embryo is not different from that found in the general population (Sill et al., 2000). This is not borne out by the data (Behr et al., 1999Go; Blickstein et al., 1999Go; Saito et al., 2000Go) which described increased rates of MZT in single replaced embryos and blastocysts. Derom et al. (1987) hypothesized that ovarian stimulation itself might cause hardening of the ZP, as observed in animals by Lungo (Lungo, 1981Go). If this hardening is not uniform, some weak spots could be formed through which the blastocyst might herniate. These authors also calculated a higher rate of monozygosity in ovulation induced triplets than in induced twins.

It is interesting that five of our cases were dizygotic triplets, whereas we identified in the same cohort of patients only two cases of monozygotic twins. Our literature search revealed 95 sets of monozygotic twins after assisted reproduction as opposed to 18 sets of dichorionic triplets or higher-order monozygotic pregnancies (Table IIIGo). The ratio of triplets to twins in nature is significantly less than that found in the setting of assisted reproduction. The mechanism responsible for the irregular distribution of MZT triplets after assisted reproduction is a matter of speculation, which warrants additional study.

Two possible biases limit our ability to draw firm conclusions from this study. Firstly, the low incidence of this phenomenon makes a large series difficult to accumulate. To be able to achieve conclusions with statistical significance, sample sizes of 10 000–20 000 cases are needed. These requirements necessitate multicentric cooperation or meta-analysis. Secondly, because we relied on the early diagnosis of pregnancy sacs and membranes to predict zygosity accurately, we might be underestimating the true incidence of MZT in cases of dichorionic biamniotic monozygotic twinning; our ultrasound criteria might mistakenly label these pregnancies as non-MZT, and accurate post-natal follow-up data on all same-sex multiple pregnancy children was not available. However, if these data are inaccurate, they only underestimate the true higher incidence of MZT.

The cumulative combined MZT rate of all published series was 1.51%, approximately three times the background rate in the untreated population (Table IIIGo). It is significant that in all series reported, the incidence of MZT was increased, regardless of its being subsequent to ovulation induction, IVF and/or micromanipulation—of any type—most often assisted hatching or ICSI (Edwards et al., 1986Go; Wenstrom et al., 1993Go; Alikani et al., 1994Go; Slotnick and Ortega, 1996Go; Hershlag et al., 1997Go; Blickstein et al., 1999Go; Saito et al., 2000Go; Sills et al., 2000Go)(Table IIIGo). This finding supports the surmise that the ovulation induction stimulus could be a major influence promoting MZT in assisted reproduction, as opposed to the current thinking implicating zonal manipulation. The basic underlying common denominator of these three situations is ovulation induction therapy, which seems to predispose either to monozygotic twinning or enhanced survival of monozygotic twins after their formation. It is possible that improved endometrial conditions after gonadotrophin therapy encourage monozygotic implantation, or that the biochemical milieu of the uterine cavity after gonadotrophin therapy encourages asymmetrical ZP hatching, independently of zonal manipulation procedures done in vitro. Each of our cases represents a different arm of assisted reproduction therapies, gonadotrophin therapy, IVF–embryo transfer and micromanipulation and embryo transfer. This leads us to the same conclusion as Derom and his associates—that gonadotrophin treatment can increase the incidence of MZT in all patients so treated (Derom et al., 1987Go). The phenomenon of MZT, although uncommon, is another factor to take into account during assisted reproduction treatment. It would appear that the dizygotic triplet combination is even more common than one would expect, respective to the monozygotic twin. Ovarian stimulation for ovulation induction or IVF–embryo transfer should be undertaken carefully, and the number of induced follicles limited, or the number of embryos replaced limited, in order to minimize multiple pregnancies. Extra consideration should be applied to account for the increased multiple pregnancy rate, partially due to an increased rate of MZT (Derom et al., 1993Go). The problems of fetal reduction are also amplified in cases of monozygosity, so that this modality may not always be available to fall back on. This short series also emphasizes the extremely poor obstetric outcome of monozygotic multiple pregnancies, another reason to exercise caution in decision-making in assisted reproduction treatment.


    Notes
 
1 To whom correspondence should be addressed at: IVF and Infertility Unit, The Department of Obstetrics and Gynecology,Assaf Harofeh Medical Center, Zerifin, Israel 70300. E-mail: ivfdoc{at}asaf.health.gov.il Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Abusheikha, N., Salha, O., Sharma, V. and Brinsden, P. (2000) Monozygotic twinning and IVF/ICSI treatment: a report of 11 cases and review of literature. Hum. Reprod. Update, 6, 396–403.[Abstract/Free Full Text]

Alikani, M., Noyes, N., Cohen, J. and Rosenwaks, Z. (1994) Monozygotic twinning in the human is associated with the zona pellucida architecture. Hum. Reprod., 9, 1318–1321.[Abstract]

Avrech, O., Schoenfeld, A., Amit, S. et al. (1993) Dizygotic triplet pregnancy following in vitro fertilization. Hum Reprod., 8, 2240–2242.[Abstract]

Biljan, M.M., Hewitt, J., Kingsland, C.R. and Taylor, C.T. (1995) Trizygotic quadruplet pregnancy following in-vitro fertilization: an additional factor against replacement of three embryos in young patients? Hum. Reprod., 10, 2169–2170.[Abstract]

Blickstein, I., Verhoeven, H.C. and Keith, L.G. (1999) Zygotic splitting after assisted reproduction. N. Engl. J. Med., 340, 738–739.[Free Full Text]

Behr, B., Fisch, J.D., Milki, A.A. et al. (1999) Blastocyst transfer is associated with an increased incidence of monozygotic twinning. The 15th Annual meeting of the European Society of Human Reproduction and Embryology, Tours, France. Hum. Reprod. 14 (Abstract Bk. 1), P-081, p. 181.

Bulmer, M.G. (1970) The biology of twinning in man. Clarendon Press, Oxford.

Cohen, J. and Feldberg, D. (1991) Effects of the size and number of zona pellucida openings on hatching and trophoblast outgrowth in the mouse embryo. Mol. Reprod. Develop., 30, 70–78.[ISI][Medline]

Derom, C., Derom, R., Vlietinck, R. et al. (1987) Increased monozygotic twinning rate after ovulation induction. Lancet, i, 1236–1238.

Derom, C., Derom, R., Vlietinck, R. et al. (1993) Iatrogenic multiple pregnancies in East Flanders, Belgium. Fertil Steril., 60, 493–496.[ISI][Medline]

Edwards, R.G., Mettler, L. and Walters, D.E. (1986) Identical twins and in vitro fertilization. J. IVF–ET, 3, 114–117.

Hershlag, A., Paine, T., Cooper, G.W. et al. (1997) Monozygotic twinning associated with mechanical assisted hatching. Fertil. Steril., 71, 144–146.

Inion, I., Gerris, J., Joostens, M. et al. (1998) An unexpected triplet heterotopic pregnancy after replacement of two embryos. Hum. Reprod., 13, 1999–2001.[Abstract]

Lungo, F.J. (1981) Changes in the zona pellucida and the plasmalemma of aging mouse eggs. Biol. Reprod., 253, 399–411.

Malter, H.E. and Cohen, J. (1989) Partial zona dissection of the human oocyte: a non traumatic method using micromanipulation to assist zona pellucida penetration. Fertil. Steril., 51, 139–148.[ISI][Medline]

Monteagudo, A., Timor-Tritsch, I. and Sharma, S. (1994) Early and simple determination of chorionic and amniotic type in multifetal gestations in the first fourteen weeks by high-frequency transvaginal ultrasonography. Am. J. Obstet. Gynecol., 170, 824–829.[ISI][Medline]

Ron-el, R., Herman, A. and Golan, A. (1991) Gonadotrophins and combined gonadotrophin releasing hormone agonist and gonadotrophin protocols in a randomized prospective study. Fertil. Steril., 55, 574–578.[ISI][Medline]

Saito, H., Tsutsumi, O., Noda, Y. et al. (2000) Do assisted reproductive technologies have effects on the demography of monozygotic twinning? Fertil Steril., 74, 178–179.[ISI][Medline]

Salat-Baroux, J., Alvarez, S. and Antoine, J.M. (1994) A case of triple monoamniotic pregnancy combined with a biamniotic twinning after in-vitro fertilization. Hum. Reprod., 9, 374–375.[ISI][Medline]

Schieve, L.A., Meikle, S.F., Peterson, H.B. et al. (2000) Does assisted hatching pose a risk for monozygotic twinning in pregnancies conceived through in vitro fertilization? Fertil. Steril., 74, 288–294.[ISI][Medline]

Slotnick, R.N. and Ortega, J.E. (1996) Monoamniotic twinning and zona manipulation: a survey of US IVF centers correlating zona manipulation procedures and high-risk twinning frequency. J. Assist. Reprod. Genet., 13, 381–385.[ISI][Medline]

Sills, E.S., Moomjy, M., Zaninovic, N. et al. (2000) Human zona pellucida micromanipulation and monozygotic twinning frequency after IVF. Hum. Reprod., 15, 890–895.[Abstract/Free Full Text]

Steiner, H.P. and Ojakangas, C.L. (1994) Dizygotic triplet pregnancy following in-vitro fertilization. Hum. Reprod., 9, 1362–1363.[ISI][Medline]

Townsend, R.R., Simpson, G.F. and Filly, R.A. (1988) Membrane thickness in ultrasound prediction of chorionicity of twin gestations. J. US Med., 7, 327–332.

Van Steirteghem, A., Nagy, Z., Joris, H. et al. (1993) High fertilization and implantation rates after intracytoplasmic sperm injection. Hum. Reprod., 7, 1061–1066.[Abstract]

Wenstrom, K.D., Syrop, C.H., Hammit, D.G. and VanVoorhis, B.J. (1993) Increased risk of monochorionic twinning associated with assisted reproduction. Fertil. Steril., 60, 510–514.[ISI][Medline]

Winn, H.N., Gabrielli, S., Reece, E.A. et al. (1989) Ultrasonographic criteria for the prenatal diagnosis of placental chorionicity in twin gestations. Am. J. Obstet. Gynecol., 161, 1540–1542.[ISI][Medline]

Submitted on June 26, 2000; accepted on March 9, 2001.