Complete Hydatidiform Mole With Coexisting Live Fetus: Outcomes After Assisted Reproductive Technology and Natural Conception-Two Case Reports

Complete hydatidiform mole with a coexisting live fetus in a twin pregnancy is exceptionally rare and carries major maternal and fetal risks. We describe two such pregnancies to illustrate management options. The first followed frozen–thawed single blastocyst transfer and presented at 13 weeks with a live fetus, a separate molar mass, an intramural fibroid, markedly elevated serum beta–human chorionic gonadotropin, and biochemical hyperthyroidism. Multidisciplinary assessment led to hysterotomy with removal of the fetus, molar tissue, and fibroids, followed by rapid recovery and no gestational trophoblastic neoplasia. The second was a spontaneously conceived twin pregnancy with a live 9‐week fetus and multicystic molar tissue that was evacuated by suction curettage. An initial decline in beta–human chorionic gonadotropin was followed by a low‐level plateau that resolved after a single low‐dose methotrexate cycle. These cases highlight the need for individualized, multidisciplinary care and close biochemical follow‐up.

assisted reproductive technology, coexisting fetus, complete hydatidiform mole, gestational trophoblastic neoplasia, twin pregnancy 1. Introduction Coexistence of a viable fetus and a complete hydatidiform mole (CHM) in a twin pregnancy, often referred to as complete hydatidiform mole with coexisting fetus (CHMCF), is an exceedingly rare phenomenon, with an estimated incidence of 1 in 20 000–100 000 pregnancies [1, 2]. The incidence appears to be higher among patients undergoing assisted reproductive technologies (ART), particularly IVF and ICSI cycles, due to multiple embryo transfers and advanced maternal age [3]. Known risk factors for molar pregnancies include extremes of maternal age, previous molar gestation, and ovulation induction [4]. Hydatidiform moles with a coexisting fetus are generally categorized into three subtypes: (i) a normal fetus with a coexisting CHM (46XX or 46XY); (ii) a normal fetus with a coexisting partial hydatidiform mole (PHM; 69XXX or 69XXY); and (iii) a triploid fetus with PHM. Diagnosis is typically established in the late first to early second trimester by ultrasonography—demonstrating a viable fetus alongside a grape‐like cystic molar mass—and confirmed by markedly elevated serum beta–human chorionic gonadotropin (β‐hCG) levels. Most cases are identified between 11 and 16 weeks of gestation [5]. Maternal complications such as heavy vaginal bleeding, early‐onset preeclampsia, hyperthyroidism, and progression to gestational trophoblastic neoplasia (GTN) are frequent. Fetal prognosis is generally poor, with high rates of miscarriage, prematurity, and growth restriction, and only a minority of pregnancies resulting in a structurally normal healthy liveborn. Consequently, management is controversial and ranges from early termination of pregnancy to expectant management with intensive maternal–fetal surveillance, and decisions are usually individualized within a multidisciplinary team [6]. Here, we present two cases of CHMCF—one following assisted reproduction and one after spontaneous conception—to illustrate the wide clinical spectrum and the need for individualized, multidisciplinary management. 2. Case Reports 2.1. Case 1 A 33‐year‐old gravida 2, abortus 1 woman presented at 13 weeks and 4 days of gestation with vaginal bleeding and lower abdominal pain. The pregnancy had been achieved following day‐5 frozen–thawed single blastocyst transfer using a vitrification and warming protocol. Her surgical history included laparoscopic left ovarian cystectomy for endometriosis. Since the beginning of pregnancy, she had been using vaginal progesterone and taking 400 μg of folic acid daily. The nuchal translucency (NT) was measured at 1.6 mm and the nasal bone was visualized during first‐trimester ultrasonographic assessment. The patient was referred to our clinic due to the identification of an intrauterine multicystic mass. Transvaginal ultrasonography revealed a single live fetus with a crown–rump length (CRL) of 70.8 mm and positive fetal cardiac activity. A grape‐like vesicular structure consistent with molar tissue was identified adjacent to the placenta. Additionally, a 7‐cm intramural fibroid was noted in the anterior wall of the lower uterine segment, extending toward and functionally obstructing the cervical canal (Figure 1). Anti‐D immunoglobulin was administered at this time because of minimal bleeding in the setting of Rh incompatibility. Initial laboratory evaluation revealed a β‐hCG level of 1 312 523 mIU/mL and signs of thyrotoxicosis (TSH < 0.005 mIU/L, free T4 59.01 pmol/L, and free T3 15.4 pmol/L), while the remainder of the workup was unremarkable. During hospitalization, the patient developed tachycardia (pulse 112 bpm) and worsening nausea and vomiting. Endocrinology consultation was obtained, and propylthiouracil (PTU) 50 mg three times daily was initiated. Thyroid autoantibodies (TSI, anti‐TPO, anti‐Tg) were negative. A multidisciplinary evaluation by endocrinology and perinatology teams concluded that the prognosis was poor due to unstable vital signs, ongoing vaginal bleeding, and hyperthyroidism. The patient was extensively counseled regarding maternal risks (massive hyperthyroidism with β‐hCG > 1.3 million mIU/mL, ongoing bleeding, and a 27%–46% risk of subsequent GTN), the guarded fetal prognosis in CHMCF (live birth rates of 29%–50% in published series), and the available management options including expectant continuation with intensive surveillance versus surgical termination. Non‐invasive prenatal testing (NIPT) was considered but deemed unreliable in this setting due to confounding cell‐free DNA contribution from the molar component, and invasive testing (chorionic villus sampling or amniocentesis) was associated with increased procedural complications in the presence of molar tissue and was therefore not pursued. After shared decision‐making, surgical termination was elected. Suction evacuation was considered but deemed technically prohibitive given the 7‐cm intramural fibroid functionally obstructing the cervical canal and the volume of molar tissue with a 14‐week‐sized fetus; expectant management was deemed unsafe given the magnitude of biochemical thyrotoxicosis (free T4≈4× upper limit of normal), the established association of β‐hCG levels above 1 000 000 mIU/mL with adverse maternal outcomes, and the inability to ensure prompt evacuation in the event of acute clinical deterioration. Hysterotomy was performed under spinal anesthesia via a Pfannenstiel incision. A complete mole and a 14‐week‐sized fetus with an intact placenta were removed. Intraoperatively, two additional fibroids were identified and excised. Thyroid hormone levels subsequently normalized. Serum hCG levels were monitored weekly until normalization (< 5 mIU/mL) and subsequently monthly for 6 months, in accordance with International Federation of Gynecology and Obstetrics (FIGO) recommendations (Figure 2). Both placentas were submitted separately for histopathological examination (Figure 3). The molar placenta (448 g, 17 × 12.5 × 4 cm) lacked an umbilical cord and consisted entirely of multilobular cystic vesicular tissue, whereas the co‐twin placenta (42 g, 7 × 5 × 3 cm) showed normal villous architecture with a centrally inserted three‐vessel umbilical cord. Microscopy of the molar placenta revealed diffuse hydropic villi with central cisternal cavitation and circumferential trophoblastic proliferation involving both cytotrophoblast and syncytiotrophoblast. Immunohistochemistry demonstrated diffuse positivity for hCG and human placental lactogen (HPL), focal positivity for placental alkaline phosphatase (PLAP), a “null” mutant pattern for p53, and loss of p57KIP2 nuclear expression in the molar placenta; in contrast, p57 staining was preserved in the cytotrophoblast and villous stromal cells of the co‐twin placenta, providing an intrinsic positive control and supporting the diagnosis of androgenetic CHM while supporting exclusion of partial mole and placental mesenchymal dysplasia. The pathology report explicitly noted that confirmatory karyotype analysis was not performed and would be appropriate for definitive molecular characterization. 2.2. Case 2 A 31‐year‐old G1P0 patient was referred to our perinatology unit at 8 weeks of gestation after a spontaneous conception. She had been taking 400 μg of folic acid daily since the beginning of her pregnancy. Ultrasonography revealed a live 8‐week fetus in one gestational sac and an area in the other sac that was consistent with a multicystic hydatidiform mole (Figure 1). Her initial β‐hCG level was 331 418 mIU/mL, and the rest of the laboratory values including thyroid function tests, were within normal limits. The patient was counseled regarding termination of pregnancy, and at 9 weeks of gestation she underwent suction aspiration and curettage under sedation. The procedure was performed without any early postoperative complications. Postoperatively, she received antibiotics and analgesics. Weekly β‐hCG monitoring showed a steady decline; however, between postoperative weeks 5 and 7 the values plateaued at 26.4, 20.0, and 21.0 mIU/mL across three consecutive weekly measurements (Figure 2). The inter‐measurement change between the final two values was +5%, well within the ≤ 10% plateau threshold. At that point, formal FIGO 2021 criteria for GTN—which require four hCG values demonstrating a plateau over 3 weeks—had not yet been fulfilled, as only three weekly measurements were available. Nonetheless, biochemical plateau was clearly established. Following multidisciplinary consensus, a single dose of methotrexate 50 mg/m2 was administered preemptively rather than awaiting a fourth measurement, balancing the recognized increased GTN risk in CHMCF against the well‐documented safety profile of low‐dose methotrexate in low‐risk post‐molar persistence. β‐hCG subsequently declined gradually (17.0, 13.3, 13.0, and 9.3 mIU/mL over the next 4 weeks) and reached undetectable levels (< 2 mIU/mL) approximately 24 days after methotrexate administration. Stable baseline values (1.2 mIU/mL) were maintained throughout more than 8 months of subsequent surveillance, and no clinical or biochemical evidence of GTN developed. Spontaneous resolution of persistent low‐level β‐hCG cannot be excluded, and a causal effect of methotrexate cannot be inferred from this single observation. Histopathological examination of the curettage material confirmed CHM, with diffuse villous hydropic change, circumferential trophoblastic proliferation, and loss of p57KIP2 nuclear expression in villous stromal cells and cytotrophoblasts on immunohistochemistry (Figure 3). The molar tissue had been collected separately from the co‐twin gestational sac at the time of suction evacuation. Molecular genotyping was not performed. 3. Discussion Meta‐analyses and systematic reviews report live birth rates in CHMCF cases between 29% and 50%, with approximately 10%–14% reaching term and a high rate of preterm delivery [1, 6, 7]. Maternal complications—including bleeding, hyperthyroidism (≈25%), hypertension, and persistent GTN—occur in up to 80% of cases [3]. The risk of GTN in coexisting molar and normal pregnancy compared with singleton molar pregnancy is increased from 15%–20% to 27%–46%. IVF/ICSI‐related pregnancies account for roughly 12%–25% of reported CHMCF cases [8]. In both cases, the diagnosis of CHM was established by a combination of histopathological criteria (diffuse villous hydropic change with cisternal cavitation and circumferential trophoblastic proliferation) and p57KIP2 immunohistochemistry. p57 is a paternally imprinted, maternally expressed cyclin‐dependent kinase inhibitor, and its loss of expression in villous stromal cells and cytotrophoblast is highly characteristic of androgenetic CHM, distinguishing it from partial hydatidiform mole (which retains p57 expression due to preserved maternal genomic contribution) and from placental mesenchymal dysplasia (which shows enlarged stem villi with cisterns but preserved p57 expression and an absence of trophoblastic proliferation) (Figure 3). In Case 1, the diagnosis was further supported by an internal control afforded by the dual‐placenta specimen: the molar placenta showed complete loss of p57 expression while the co‐twin placenta exhibited preserved p57 staining in cytotrophoblast and villous stromal cells, together with a full immunohistochemical panel (hCG, HPL, p53, PLAP). Although short tandem repeat (STR) genotyping or ploidy analysis would represent the gold standard for definitive confirmation of androgenetic diploidy, these molecular studies were not performed at our institution and this represents a limitation of our diagnostic workup. The pathology report for Case 1 explicitly noted that confirmatory karyotype analysis was not available. In the first case, the NT measurement of 1.6 mm and visualization of the nasal bone during first‐trimester screening suggested a low risk of chromosomal aneuploidy for the morphologically normal fetus, although first‐trimester sonographic markers alone cannot reliably exclude triploidy or other ploidy abnormalities in this setting. Nevertheless, in twin pregnancies with CHMCF, even when the fetus appears morphologically normal, the molar placental component raises concern for potential triploidy or other chromosomal abnormalities. Invasive diagnostic procedures such as chorionic villus sampling (CVS) or amniocentesis carry an increased risk of complications due to the presence of molar tissue and should therefore be considered with caution [9]. In both of our patients, we did not perform a diagnostic test due to a decision for termination of pregnancy. Furthermore, the first case's pregnancy has been achieved after a single embryo transfer. This case illustrates that single embryo transfer does not preclude the occurrence of CHMCF. The precise embryologic mechanism underlying CHMCF after single embryo transfer remains uncertain; proposed possibilities include monozygotic splitting of the transferred embryo with a concurrent abnormal fertilization event, although such mechanisms remain speculative and were not directly demonstrated in our case [10]. Predictors of pregnancy continuation and favorable outcome include the absence of antenatal complications, β‐hCG levels < 400 000 mIU/mL at diagnosis, and normal thyroid function and blood pressure [2]. Conversely, β‐hCG levels above 1 000 000 mIU/mL and abnormal thyroid function tests are associated with adverse outcomes. The phenomenon known as the “hook effect” can lead to falsely low β‐hCG measurements in cases of extremely high trophoblastic burden. After sample dilution, Levavi et al. reported extreme β‐hCG values of 1.6 million and 2.2 million mIU/mL in molar pregnancies which represent some of the highest reliably measured β‐hCG levels described in the literature [11]. The level of 1 312 523 mIU/mL in the first case, therefore, also ranks among the highest reliably measured values reported in CHMCF. In the second case, despite initially high β‐hCG levels, an appropriate early decline was observed as expected; however, a subsequent plateau at low positive levels indicated persistent trophoblastic activity. Although FIGO criteria for GTN were not strictly met, such biochemical persistence is clinically important. Low‐dose methotrexate has been reported to rapidly normalize β‐hCG in cases of post‐molar biochemical persistence, although robust evidence for true prevention of subsequent GTN remains limited and derives largely from small series and single‐arm studies [12]. In our second case, complete biochemical remission followed a single methotrexate dose; however, spontaneous resolution cannot be excluded and a causal effect of methotrexate cannot be inferred from a single observation. Existing series do not clearly demonstrate that termination of pregnancy reduces the risk of GTN compared with expectant management in CHMCF [13]. Older cohorts even suggest comparable or higher rates of persistent GTN in pregnancies terminated before viability, whereas a recent review of 42 cases found an overall GTN incidence of around 40% without a clear relationship to gestational age at termination [14]. Importantly, the risk of progression differs markedly by mole type: while complete moles carry a 15%–20% risk of post‐molar neoplasia, partial moles carry a risk of only around 1.5%, and expectant management of partial mole with a coexisting fetus has been reported to result in term delivery of a healthy neonate (Libretti et al.) [15]. This distinction is clinically relevant, as favorable outcomes described for partial mole cannot be extrapolated to CHMCF, where both maternal and oncologic risks are substantially higher. Thus, GTN risk appears to remain high regardless of management choice, and decisions should primarily be guided by maternal–fetal condition and patient preferences. The optimal delivery method after expectant management in CHMCF remains a matter of debate. A 2021 meta‐analysis of 72 CHMCF cases reported that 72.2% of patients underwent cesarean section and 27.8% delivered vaginally, with no significant difference in post‐molar GTN incidence between the two groups [2]. This suggests that vaginal delivery may be a safe option when obstetrically appropriate. For first‐trimester molar evacuation, suction curettage is preferred because it offers shorter operative time and more effective evacuation compared with sharp curettage, without any significant difference in GTN incidence or major complications between the techniques [16]. However, in pregnancies beyond 16 weeks of gestation or in the presence of complicating factors such as large fibroids or unstable maternal status, hysterotomy may be indicated, as in our first case. Thus, in CHMCF with a live fetus, vaginal delivery can be considered in stable situations when obstetrically appropriate. However, additional risk factors must be carefully evaluated and the route of delivery should be decided only after thorough counseling and shared decision‐making with the patient. In conclusion, these two cases illustrate the broad clinical spectrum of CHMCF and emphasize the need for individualized, multidisciplinary management. Extremely high β‐hCG levels and maternal complications may justify early termination of pregnancy, whereas persistent low‐level positivity after evacuation may prompt individualized consideration of low‐dose chemotherapy in selected patients, with the recognition that spontaneous resolution is also possible. Careful maternal–fetal monitoring and close biochemical follow‐up remain essential to minimize morbidity. Author Contributions Mert Eyupoglu: conceptualization, writing – original draft, writing – review and editing, visualization. Diclehan Orhan: visualization, writing – review and editing. Erdem Fadiloglu: conceptualization, project administration, writing – review and editing, methodology. Muhammed Onur Atakul: investigation, writing – original draft. Ozgur Deren: writing – review and editing, supervision. Murat Cagan: visualization, writing – review and editing, writing – original draft, supervision. Disclosure The authors have nothing to report. Ethics Statement This case report was exempt from institutional review board approval in accordance with the institutional policy of Hacettepe University Hospital, Faculty of Medicine, regarding case reports. The study was conducted in accordance with the principles of the Declaration of Helsinki. Consent Written informed consent for publication of the case details and accompanying images was obtained from both patients. Conflicts of Interest The authors declare no conflicts of interest. Acknowledgments The authors have nothing to report. Data Availability Statement Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

- 1. Zilberman Sharon N., Maymon R., Melcer Y., and Jauniaux E., “Obstetric Outcomes of Twin Pregnancies Presenting With a Complete Hydatidiform Mole and Coexistent Normal Fetus: A Systematic Review and Meta‐Analysis,” BJOG: An International Journal of Obstetrics and Gynaecology 127, no. 12 (2020): 1450–1457. [DOI] [PubMed] [Google Scholar] - 2. Wang G., Cao J., Xu X., Han X., and Cui H., “Delivery Management of a Complete Hydatidiform Mole and Co‐Existing Viable Fetus: A Meta‐Analysis and Systematic Review,” Journal of Gynecology Obstetrics and Human Reproduction 51 (2021): 102269. [DOI] [PubMed] [Google Scholar] - 3. Wang G., Cui H., and Chen X., “A Complete Hydatidiform Mole and Coexisting Viable Fetus in a Twin Pregnancy: A Case Report With Literature Review,” Journal of Maternal‐Fetal & Neonatal Medicine 36, no. 1 (2023): 2183746. [DOI] [PubMed] [Google Scholar] - 4. Sebire N. J., Foskett M., Paradinas F. J., et al., “Outcome of Twin Pregnancies With Complete Hydatidiform Mole and Healthy Co‐Twin,” Lancet 359, no. 9324 (2002): 2165–2166. [DOI] [PubMed] [Google Scholar] - 5. Sood P., Gupta J., and Tholia R., “Complete Hydatidiform Mole With a Coexisting Twin Live Fetus (CHMTF): The Uncommon Diagnostic Enigma—Simplified,” Egyptian Journal of Radiology and Nuclear Medicine 55, no. 1 (2024): 77. [Google Scholar] - 6. Mok Z. W., Merchant K., and Yip S. L., “Management of a Complete Hydatidiform Mole With a Coexisting Live Fetus Followed by Successful Treatment of Maternal Metastatic Gestational Trophoblastic Disease: Learning Points,” BML Case Reports 14, no. 1 (2021): e235028. [DOI] [PMC free article] [PubMed] [Google Scholar] - 7. Hajri T., Massoud M., Vergne M., et al., “Multiple Pregnancy With Complete Hydatidiform Mole and Coexisting Normal Fetus in a Retrospective Cohort of 141 Patients,” American Journal of Obstetrics and Gynecology 230, no. 3 (2024): 362.e1–362.e8. [DOI] [PubMed] [Google Scholar] - 8. Ngan H. Y. S., Seckl M. J., Berkowitz R. S., et al., “Diagnosis and Management of Gestational Trophoblastic Disease: 2021 Update,” International Journal of Gynecology & Obstetrics 155, no. S1 (2021): 86–93. [DOI] [PMC free article] [PubMed] [Google Scholar] - 9. Jauniaux E., Memtsa M., Johns J., Ross J. A., and Jurkovic D., “New Insights in the Pathophysiology of Complete Hydatidiform Mole,” Placenta 62 (2018): 28–33. [DOI] [PubMed] [Google Scholar] - 10. Niemann I., Bolund L., and Sunde L., “Twin Pregnancies With Diploid Hydatidiform Mole and Co‐Existing Normal Fetus May Originate From One Oocyte,” Human Reproduction 23, no. 9 (2008): 2031–2035. [DOI] [PubMed] [Google Scholar] - 11. Levavi H., Neri A., Bar J., Regev D., Nordenberg J., and Ovadia J., ““Hook Effect” in Complete Hydatidiform Molar Pregnancy: A Falsely Low Level of Beta‐HCG,” Obstetrics and Gynecology 82, no. 4 Pt 2 Suppl (1993): 720–721. [PubMed] [Google Scholar] - 12. Biswas J., Dasgupta S., Datta M., Datta M., Saha S., and Pradhan P., “Effect of Single‐Dose Methotrexate Injection to Prevent Neoplastic Changes in High Risk Complete Hydatidiform Mole: A Randomised Control Trial,” Journal of Family Medicine and Primary Care 11, no. 10 (2022): 6036–6041. [DOI] [PMC free article] [PubMed] [Google Scholar] - 13. Salmeri N., Pizzetti A., Grassi E., et al., “Multiple Pregnancy With Complete Hydatidiform Mole and Coexisting Normal Fetus: Systematic Review and Meta‐Analysis of Clinical Outcomes From Non‐Randomized Studies,” Ultrasound in Obstetrics & Gynecology 67, no. 3 (2026): 272–282, 10.1002/uog.70104. [DOI] [PMC free article] [PubMed] [Google Scholar] - 14. Liuqin H., Na W., Huimin Z., et al., “Complete Mole With Co‐Existing Fetus: A Rare Case Report of Full‐Term Delivery With Favorable Maternal and Fetal Outcomes,” Taiwanese Journal of Obstetrics & Gynecology 64, no. 6 (2025): 1056–1060. [DOI] [PubMed] [Google Scholar] - 15. Libretti A., Longo D., Faiola S., De Pedrini A., Troìa L., and Remorgida V., “A Twin Pregnancy With Partial Hydatidiform Mole and a Coexisting Normal Fetus Delivered at Term: A Case Report and Literature Review,” Case Reports in Women's Health 39 (2023): e00544. [DOI] [PMC free article] [PubMed] [Google Scholar] - 16. Sato A., Usui H., and Shozu M., “Comparison Between Vacuum Aspiration and Forceps Plus Blunt Curettage for the Evacuation of Complete Hydatidiform Moles,” Taiwanese Journal of Obstetrics & Gynecology 58, no. 5 (2019): 650–655. [DOI] [PubMed] [Google Scholar] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Data Availability Statement Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.