Twin pregnancy with a complete hydatidiform mole and co-existing fetus following in-vitro fertilization: Case report

Francisco Montes-de-Oca-Valero1, Lena Macara1,3 and Adel Shaker2

1 The Queen Mother's Hospital, Yorkhill, Glasgow G3 8SJ and 2 Ross Hall Hospital, Glasgow, UK


    Abstract
 Top
 Abstract
 Introduction
 Case report
 Discussion
 References
 
Hydatidiform mole with a co-existing live fetus is a rare event. We report the case of a 41 year old Caucasian woman who attended for in-vitro fertilization. Three embryos, containing two apparently normal pronuclei, were transferred into the uterus. A twin pregnancy with a live fetus and a complete mole ensued. The pregnancy was managed conservatively until 28 weeks gestation when, following an episode of major antepartum haemorrhage, a live female infant was delivered by Caesarean section. The mole, weighing over 1.7 kg, was successfully evacuated. Following delivery, serum human chorionic gonadotrophin concentrations returned to baseline and remain within the normal range at 24 months. Both mother and daughter are well on assessment 24 months later.

Key words: HCG/in-vitro fertilization/molar pregnancy/placenta/twin pregnancy


    Introduction
 Top
 Abstract
 Introduction
 Case report
 Discussion
 References
 
Gestational trophoblastic disease is known to be associated with increased maternal age and is more commonly seen in Far East populations. In the late 1970s, the single disease entity was subdivided into partial moles with a fetal pole, often with triploidy, and complete molar pregnancies, without fetal tissue, which were diploid but derived entirely from paternal genes (Vassilakos et al., 1977Go; Szulman and Surti 1978aGo,bGo). Rarely, even in the Far East, a complete hydatidiform mole (CHM) and a co-existing normal pregnancy is found (Harada et al., 1997Go; Chen, 1997Go). When the diagnosis has been made ante-natally, most patients have opted for termination of pregnancy to avoid/reduce the risks of malignant change and complications such as pre-eclampsia and antepartum haemorrhage (APH) (Steller et al., 1994Go). However, where CHM and a co-existing fetus occurs in a woman achieving a first conception in later life or after many attempts of assisted conception, the dilemma is enormous. We report such a case, diagnosed in the second trimester which, in spite of the molar size, was managed conservatively with a successful pregnancy outcome for both mother and baby.


    Case report
 Top
 Abstract
 Introduction
 Case report
 Discussion
 References
 
This 41 year old, para 1 + 0 was known to have endometriosis. Her first pregnancy was a spontaneous conception after a lengthy time of primary infertility and was uncomplicated. After 7 years of secondary infertility she was referred for assisted conception. Following ovarian suppression with gonadotrophin-releasing hormone (GnRH) analogue, follicular growth was stimulated using purified urinary menopausal gonadotrophins. Fourteen oocytes were retrieved under transvaginal ultrasound guidance and the oocytes then inseminated in a concentration of 100 000 motile spermatozoa per oocyte. Fertilization was confirmed the following day and eight oocytes noted to be normally fertilized (two pronuclei seen). These all cleaved and three 4-cell embryos (grade 1) were transferred into the uterus 48 h after egg collection. Progesterone pessaries were used for luteal support. Four weeks following embryo transfer two fetal poles, each with a positive fetal heart, were identified on transvaginal ultrasound. In addition, a third poorly defined sac was outlined. The patient had recurrent episodes of fresh vaginal bleeding but since repeat ultrasound demonstrated only one viable fetal pole the bleeding was ascribed to failure of the other pregnancies. By 16+ weeks gestation the `collapsed' sac had increased substantially in size and appeared cystic. An elevated serum human chorionic gonadotrophin (HCG) level of 22.77 multiples of the medium (MOM) (840 000 mIU/ml) was noted on second trimester serum screening. The patient was referred for further evaluation.

Ultrasound demonstrated a viable, structurally normal 16 week old fetus with a normal-looking placenta. The entire lower segment of the uterus was occupied by a large cystic mass. The appearance was in keeping with a molar pregnancy, which was significantly larger than the normal pregnancy (Figure 1Go). During the following 10 weeks, the patient had several episodes of APH, all of which settled spontaneously. In addition, she also developed significant proteinuria with mild hypertension and borderline renal function. The `mole' continued to grow in size with advancing gestation and the uterus remained very large for dates. Intramuscular betamethasone was administered to promote fetal lung maturity. At 27 + 6 weeks gestation, the patient had a significant APH and continued to bleed heavily. As the mole occupied the entire lower segment, she was delivered by classical Caesarean in the maternal interest. A live female infant (980 g) and normal placenta (127 g) were delivered. Thereafter more than 1700 g of cystic, vesicular tissue was extracted and the uterus completely evacuated (Figure 2Go). Histological examination confirmed the clinical impression of one normal placenta and a second complete hydatidiform mole (46,XX). To confirm the histological diagnosis of CHM, DNA was extracted from the molar tissue and compared with DNA obtained from paternal and maternal blood lymphocytes. Polymorphic markers from six different chromosomes, 4, 6, 7, 13, 14 and 15 were examined using polymerase chain reaction (PCR). This indicated isodisomy, thus establishing that the mole pregnancy arose from a single spermatozoa (Figure 3Go).



View larger version (183K):
[in this window]
[in a new window]
 
Figure 1. Transabdominal ultrasound image of the uterus illustrating the vesicular tissue (Mo) in the lower pole sac and the normal placenta (P) in the upper second sac.

 


View larger version (142K):
[in this window]
[in a new window]
 
Figure 2. Photograph of the large vesicular mole (Mo) and the normal placenta (P) illustrating the size and vesicular nature of the mole.

 


View larger version (89K):
[in this window]
[in a new window]
 
Figure 3. A PCR gel of DNA from the molar tissue (Mo) and the maternal (M) and paternal (P) blood lymphocytes using chromosome 15 markers [MS 14 (D15597) GenBank M87695]. The molar tissue is seen to contain two copies of the paternal alleles.

 
Post-operatively, serial serum HCG concentrations showed a steady fall (Figure 4Go) and remained normal at 24 months post delivery. The infant spent 7 weeks in the paediatric unit. She developed a moderate degree of retinopathy due to prematurity, which required laser treatment, but to date has achieved normal milestones of development.



View larger version (14K):
[in this window]
[in a new window]
 
Figure 4. Serial measurements of serum human chorionic gonadotrophin (HCG) concentrations ante-natally (AN) and following delivery (P/N), demonstrating the fall in values even prior to delivery. The first P/N value was 1634 mIU/ml. Urine HCG measurements were used after 12 weeks and for follow-up thereafter. These remain in the normal range to date.

 

    Discussion
 Top
 Abstract
 Introduction
 Case report
 Discussion
 References
 
Twin pregnancies with a molar pregnancy and a co-existing viable fetus have been described (Jinno et al., 1994Go). Until recently, it was commonly a retrospective diagnosis, and therefore no clinical decision was required. Of the CHM and twins detected ante-natally, most were terminated once the diagnosis was made. As a result the natural history of pregnancies affected by a true complete mole and a co-existing twin remains unclear and the risk of other complications, particularly malignancy, is poorly defined. With the ready availability of detailed ultrasound it is likely that the number of CHM and co-existing twin pregnancies diagnosed ante-natally will increase, raising clinical dilemmas, particularly when the pregnancy has been long awaited.

It is difficult to explain how a CHM and co-existing normal twin pregnancy developed during in-vitro fertilization (IVF) treatment, when three apparently normal 4-cell embryos were replaced. Although two pronuclei were seen on day 1, there are no specific tests available which can identify the male and the female pronuclei separately. What appeared to be a second normal female `pronucleus' on day 1 after insemination may have been an empty, large second polar body. Though the second polar body is usually small and extruded at the time of fertilization, if retained it can appear as large as a normal pronucleus. Older mothers, like our patient, are known to be at increased risk of problems during meiosis, due to ageing of the oocytes. The fertilizing spermatozoon, if diploid, would have led to appropriate cleavage on day 2, giving a normal appearance to the embryo. It is well recognized that some spermatozoa remain diploid, particularly as paternal age increases. Alternatively, the two pronuclei seen could both have been male pronuclei as a result of dispermy. The molecular techniques applied, however, allowed us to verify the finger-prints of each chromosome examined. While it is conceivable that two separate spermatozoa could share a degree of homology, the likelihood of two spermatozoa sharing identical chromosomes with identical chromosome segments is extremely remote. PCR proved a useful tool to confirm that this molar pregnancy did indeed arise from a single paternal spermatozoa and was not the result of dispermy.

It is most likely that some of the clinical factors which precipitated IVF treatment for this couple (advanced maternal/paternal age, poor oocyte quality) predisposed them to a problem such as this occurring in the embryo. As many women now delay conception until later life and subsequently require assisted techniques such as IVF to achieve a successful pregnancy, this complication may well become more common. Having established an ante-natal diagnosis of CHM and co-existing twin pregnancy, counselling this patient, who desperately wished for a successful pregnancy without compromising her own life, proved difficult. Recent reviews indicate that there is a greater risk of developing persistent trophoblastic disease (PTD) (50–60%) in cases of CHM and a co-existing fetus compared with a single molar pregnancy (14%) and that when present, it more commonly progresses to metastatic disease (Steller et al., 1994Go). To date no mothers have died as a result of these complications (Fishman et al., 1998Go). However, many of these cases were evacuated at the time of diagnosis in the second trimester, and in those who continued, the moles were small in comparison with the normal pregnancy. We were faced with a mole which even on ultrasound was substantially larger than the normal fetus and placenta together.

The small number of women who have opted to continue with the pregnancy, or those who have had the diagnosis made in the third trimester, appear to have no added risk of malignant complications (Bristow et al., 1996Go), leading the authors to conclude that `advanced gestational age is not an independent risk factor for the development of PTD'. It would appear that some pregnancies with a CHM and normal twin develop malignancy irrespective of the time the pregnancy is interrupted, and even when intervention occurs in early pregnancy (Harada et al., 1997Go), the risk of malignancy persists. Ideally a test which could predict the malignant potential of the mole is required. We were able to obtain serial HCG measurements in our patient and although the initial values were extremely high, they gradually fell during the second and third trimester, despite apparent growth of the mole on ultrasound. It may be that a falling trend in serum HCG reflects less proliferation of the mole and is a good prognostic sign but it is only as more of these pregnancies are accurately documented that we will be able to confirm or refute these impressions. Given that this mole is the largest documented with a co-existing twin, malignant change would have been expected. One of the few successful cases previously reported was from China (Chen, 1997Go) and went to term without complication, subsequently developed PTD, even though the mole was completely removed at Caesarean section and was significantly smaller in size than the normal placental tissue. The quantity of molar tissue and the gestation at the time of diagnosis are not, therefore, reliable indicators of the malignant potential of a mole.

This case demonstrates two main features. Firstly, even in Caucasian women, despite apparently normal embryos being replaced during IVF, molar pregnancies may still occur. It is unlikely that this complication is directly related to the technique of IVF itself but rather that the characteristics of women and their partners attending for assisted conception make them more at risk of molar pregnancies. Secondly, even when the mole is huge, a successful outcome, without evidence of malignancy, is possible for both mother and baby.


    Notes
 
3 To whom correspondence should be addressed Back


    References
 Top
 Abstract
 Introduction
 Case report
 Discussion
 References
 
Bristow, R.E., Shumway, J.B., Khouzami, A.N. et al. (1996) Complete hydatidiform mole and surviving co-existent twin. Obstet. Gynaecol. Survey., 51,705–709.[Medline]

Chen, F-P. (1997) Molar pregnancy and living normal fetus coexisting until term: prenatal biochemical and sonographic diagnosis. Hum. Reprod., 12, 853–857.[Abstract]

Fishman, D.A., Padilla, L.A., Ken, P. et al. (1998) Management of twin pregnancy consisting of a complete hydatidiform mole and a normal fetus. Obstet. Gynaecol., 91, 546–550.[Abstract/Free Full Text]

Harada, I., Sutsumi, T., Takai, Y. et al. (1997) DNA polymorphism analysis of a case of complete hydatidiform mole co-existing with a fetus. Hum. Reprod., 12, 2563–2566.[Abstract]

Jinno, M., Ubukata, Y., Hanyu, I. et al. (1994) Hydatidiform mole with a surviving coexistent fetus following in-vitro fertilization. Hum. Reprod., 9, 1170–1172.[Abstract]

Steller, M.A., Genest, D.R., Bernstein, M.R. et al. (1994) Natural history of twin pregnancy with complete hydatidiform mole and co-existing fetus. Obstet. Gynaecol., 83, 35–42.[Abstract]

Szulman, A.E. and Surti, U. (1978a) The syndromes of hydatifiorm mole I. Cytogenic and morphologic correlations. Am. J. Obstet. Gynaecol., 131, 665–671.[ISI][Medline]

Szulman, A.E. and Surti, U. (1978b) The syndromes of hydatifiorm mole II. Morphologic evolution of the complete and partial mole. Am. J. Obstet. Gynaecol., 132, 20–27.[ISI][Medline]

Vassilakos, P., Riotton, G. and Kajii, T. (1977) Hydatidiform mole: two entities. A morphologic and cytogenetic study with some clinical considerations. Am. J. Obstet. Gynaecol., 127, 167–170.[ISI][Medline]

Submitted on March 12, 1999; accepted on July 22, 1999.