Incomplete abortion with elevated beta-human chorionic gonadotropin levels mimicking a molar pregnancy: A case report : Medicine: Case Reports and Study Protocols

Secondary Logo

Journal Logo

Clinical Case Report

Incomplete abortion with elevated beta-human chorionic gonadotropin levels mimicking a molar pregnancy

A case report

Li, Pei-Chen MDa; Chang, Kai-Hung MDa; Ding, Dah-Ching MD, PhDa,b,∗

Author Information
Medicine Case Reports and Study Protocols 2(9):p e0134, September 2021. | DOI: 10.1097/MD9.0000000000000134
  • Open


1 Introduction

Miscarriage, also known as spontaneous abortion and pregnancy loss, is the most frequent complication of pregnancy, affecting 17% to 22% of all pregnancies.[1] The clinical manifestations of miscarriage are vaginal bleeding usually with abdominal pain. Incomplete abortion is characterized by the retention of parts of the product of conception (POC) in the uterus, and the whole or a part of the placenta, fetus, or membranes may be retained. It commonly results in hemodynamic instability due to massive hemorrhage.[2]

Gestational trophoblastic disease (GTD) should be considered, especially in the presence of an abnormally elevated serum human chorionic gonadotropin (hCG) level or a suggestive pelvic ultrasound.[3] Hydatidiform mole (HM), also referred to as molar pregnancy or a mole, is the most common part of a group of GTD. It is subdivided into complete mole and partial mole. They are different on the basis of karyotype, pathology, clinical presentation, and outcome.[4]

The hCG level is 112855± 106610 and 43840 ± 46906 mIU/mL in normal first trimester pregnancy and abortion, respectively.[5] The hCG level in complete mole is usually > 100,000 mIU/mL.[6] In partial mole, the hCG level has a wide range, similar to that in normal pregnancy.[6]

Herein, we report a case of incomplete abortion with elevated beta-human chorionic gonadotropin (beta-Hcg) levels mimicking a molar pregnancy.

2 Case report

The patient was a 29-year-old woman, G6P3SA1AA2 (the nearest AA1 was in November 2017), with a past history of thalassemia and mitral valve prolapse, who presented with persistent vaginal bleeding for one month. There was no involuntary body weight loss, poor appetite, or family history of gynecological cancer. However, her mother had a history of two molar pregnancies.

She had an artificial abortion last year due to the intake of a teratogenic medication for vertigo. Thereafter, she had usual health status until 2 months ago. She experienced menorrhagia with blood clots, which was different from the previous periods. Dyspnea on exertion, dizziness, and nausea also developed during this period. She visited a local medical clinic where her hemoglobin (Hb) level was 7.6 g/dL and she received a blood transfusion. Transvaginal ultrasonography revealed a normal uterus and adnexa. Oral progesterone and tranexamic acid were prescribed for 1 week. Vaginal bleeding subsided gradually. However, after the medication was discontinued, one huge blood clot, approximately 10 cm in diameter, was expelled from her vagina, followed by massive vaginal bleeding with blood clots. Progressive exertional dyspnea and dizziness also occurred. Therefore, she visited our hospital.

At the outpatient department, the urine pregnancy test was positive. On transvaginal ultrasound, an enlarged uterus of 9.6 cm × 5.7 cm with a hematometra of 4.5 cm × 4.3 cm, thickened endometrium, and mild ascites were detected (Fig. 1A). The placenta without cystic structures was present at the posterior wall. The fetal structure was not observed, and the bilateral adnexa appeared normal. Pelvis computed tomography showed an enlarged uterus with areas of a hypoattenuating lesion (6.0 cm × 4.2 cm) surrounding the uterine cavity (Fig. 1B). Laboratory findings included an elevated serum beta-hCG level of 32980.4 mIU/mL and microcytic anemia with an Hb level of 7.2 g/dL and a mean corpuscular volume of 74.1 fl. Chest radiography revealed no active lung lesion.

Figure 1:
Ultrasound of the uterus before and after dilation and curettage. (A) An enlarged uterus of 9.6 cm × 5.7 cm with a hematometra of 4.5 cm × 4.3 cm, thickened endometrium, and mild ascites. (B) An enlarged uterus with areas of a hypoattenuating lesion (6.0 cm × 4.2) cm surrounding the uterine cavity. (C) Normal uterine appearance at 1 month postoperatively.

The patient underwent dilation and curettage (D&C). Both gestational tissue and maternal blood samples were sent for cytogenetic analysis. The final pathology revealed gestational tissue with chorionic villi, trophoblasts, and decidual tissue. Chromosome analysis of the placental tissue showed 46, XY karyotype. No chromosomal abnormality was detected in both the placental tissue and maternal blood. Therefore, we concluded this case was of an incomplete abortion.

The patient was observed closely with serial monitoring of beta-hCG levels; beta-hCG had declined to 11130.1 mIU/mL on postoperative day 1 and 89.6 mIU/mL at 1 month postoperatively. The beta-hCG level returned to normal after 3 months of follow-up and ultrasound showed a normal uterine appearance (Fig. 1C).

3 Discussion

The differential diagnosis of vaginal bleeding includes pregnancy- and hormone-related vaginal bleeding.[7] If the pregnancy test is positive, abortion or ectopic pregnancy should be considered.[8] Our case presented with vaginal bleeding and was treated with hormone and hemostasis agents, which delayed the diagnosis. Intrauterine hematoma then formed and was confused with a molar pregnancy. Thus, in women of reproductive age, a pregnancy test should be performed first to rule out the possibility of pregnancy.

Incomplete abortion can cause prolonged vaginal bleeding lasting 2 or more weeks. Delayed diagnosis or intervention leads to uterine infection and anemia and requires further blood transfusion or intervention. A study reported that 12% of women with first trimester vaginal bleeding had a spontaneous abortion, but 13% of those without bleeding experienced a miscarriage.[1] Therefore, when a patient with vaginal bleeding has a positive pregnancy test, transvaginal ultrasound and quantitative serum hCG evaluation should be performed.[9] In cases of incomplete abortion, pelvic ultrasound usually shows some POC in the uterus. The hCG level is low and no fetal heart beat is noted.[10] A per vaginal examination usually shows a large and soft uterus. Other laboratory tests include complete blood count, coagulation profile, cross match, type, and Rh factor.[9] Our case presented with long-term vaginal bleeding despite medication. The urine pregnancy test was positive and transvaginal ultrasound showed some POC in the uterus. The hCG level was 32980.4 mIU/mL and Hb was 7.2 g/dL. Incomplete abortion or molar pregnancy was then suspected.

GTD in early pregnancy may present as missed or incomplete abortion, with or without vaginal bleeding.[11,12] In early gestation, marked hydropic and cystic change may be absent on pathology and ultrasonographic examination. Therefore, the POC can be indistinguishable from a non-molar pregnancy.[12] Berkowitz et al. reported that in a series of 81 patients with partial mole, clinical diagnosis prior to uterine evacuation was incomplete or missed abortion in 74 patients (91%) and molar pregnancy in only 5 patients (6%).[13] Therefore, the differential diagnosis between incomplete abortion and molar pregnancy is important.

Molar pregnancy should be suspected by excluding non-molar pregnancy (normal intrauterine pregnancy, ectopic pregnancy, and abortion).[6] Pelvic examination, hCG evaluation, and pelvic ultrasound can help in the diagnosis of molar pregnancy. A per vaginal examination can detect a larger uterus than that in normal pregnancy.[6] Extremely high levels of serum beta-hCG are associated with a molar pregnancy, and the beta-hCG levels in partial mole are generally lower than those in complete mole. Few cases (6%) with partial mole had pre-evacuation hCG levels of > 100,000 mIU/mL.[13] Our case also presented with long-term vaginal bleeding and high beta-hCG levels, thus possibly mimicking a molar pregnancy.

In terms of ultrasonography of partial mole, the sonographic findings are enlarged placenta; cystic spaces within the placenta, which may not always be present; gestational sac, either empty or containing amorphous fetal echoes; and an embryo or fetus.[14] Ultrasonography is more difficult for diagnosing partial mole than complete mole, with an overall sensitivity of 20% and 95%, respectively.[14] Therefore, the diagnosis of partial mole mainly relies on a histologic review of curettage specimens. In our case, ultrasound also showed a hypoechoic space along the placenta, mimicking a molar pregnancy.

The final diagnosis of molar pregnancy should be made by histology, karyotype, and the presence of a fetus.[6] The most common and benign form of GTD is a HM. HM is the result of an abnormality in egg fertilization causing abnormal chorionic villi with trophoblastic hyperplasia, composed of complete mole (diploid) and partial mole (triploid, fetus present). A cytogenetic study can be subsequently performed to rule out gestational trophoblastic neoplasia. In our case, the histopathology and chromosomal studies showed normal gestational tissue and normal karyotype, which were compatible with a non-molar pregnancy.

The treatments of incomplete abortion can be surgical (D&C) or medical (misoprostol) uterine evacuation.[15] Misoprostol is considered a safe and easily accessible alternative to surgical evacuation and further prevention from subsequent surgical complications, including adhesions, uterine perforation, infections, and increased risks of an ectopic pregnancy.[16] Previous studies have shown equivalent efficacy of misoprostol and surgical curettage; therefore, the choice of treatment depends mainly upon patient preference.[15] However, in our case, tissue proof was needed to rule out molar pregnancy. Therefore, D&C was performed.

For patients with abnormal vaginal bleeding presumed secondary to endometrial dysfunction, a study found that the levonorgestrel intrauterine system (71%–95% reduction), estrogen–progestin contraceptives (35%–69% reduction), extended cycle oral progestins (87% reduction), tranexamic acid (26%–54% reduction), and non-steroidal anti-inflammatory drugs (10%–52% reduction) were all effective treatments for the reduction of menstrual blood loss.[17] Our patient was not suspected to be pregnant at first at a local clinic, and the medication was administered, thus improving the symptoms. However, this led to a delayed diagnosis of incomplete abortion.

Our case report emphasizes the following points. The first step in the management of abnormal vaginal bleeding is to determine whether the patient is hemodynamically stable. Then, a urine pregnancy test (hCG) should be performed in any woman of childbearing age, unless pregnancy is already confirmed. However, even if the pregnancy test is positive, the volume or pattern of bleeding does not confirm a spontaneous abortion. Transvaginal ultrasound and quantitative serum hCG evaluation should also be performed.

In conclusion, incomplete abortion with elevated beta-hCG levels is a rare condition that mimics a molar pregnancy. Careful preoperative evaluation of vaginal bleeding should be performed. Transvaginal ultrasound, quantitative serum hCG evaluation, and other laboratory tests should also be performed (Fig. 2).

Figure 2:
Flow chart of diagnosis for premenopausal vaginal bleeding. CBC: complete blood count, D& C: dilation and curettage, hCG: human chorionic hormone, POC: product of conception, Unit of hCG: mIU/mL.

Author contributions

PCL: study design, data interpretation, and manuscript preparation; KHC: data analysis and manuscript preparation; DCD: study design, data analysis, data interpretation, manuscript preparation, and manuscript revision. All authors agreed with the final version of the manuscript.

Conceptualization: Dah-Ching Ding.

Data curation: Pei-Chen Li, Kai-Hung Chang, Dah-Ching Ding.

Formal analysis: Pei-Chen Li, Kai-Hung Chang, Dah-Ching Ding.

Funding acquisition: Dah-Ching Ding.

Investigation: Pei-Chen Li, Dah-Ching Ding.

Methodology: Pei-Chen Li, Dah-Ching Ding.

Validation: Dah-Ching Ding.

Visualization: Dah-Ching Ding.

Writing – original draft: Pei-Chen Li, Kai-Hung Chang, Dah-Ching Ding.

Writing – review & editing: Pei-Chen Li, Dah-Ching Ding.


[1]. Hasan R, Baird DD, Herring AH, et al. Patterns and predictors of vaginal bleeding in the first trimester of pregnancy. Ann Epidemiol 2010;20:524–31.
[2]. Oliver A, Overton C. Diagnosis and management of miscarriage. Practitioner 2014;258:25–8. 3.
[3]. Shaaban AM, Rezvani M, Haroun RR, et al. Gestational trophoblastic disease: clinical and imaging features. RadioGraphics 2017;37:681–700.
[4]. Froeling FEM, Seckl MJ. Gestational trophoblastic tumours: an update for 2014. Curr Oncol Rep 2014;16:408.
[5]. Ehrentraut S, Sauss K, Neumeister R, et al. Human miscarriage is associated with dysregulations in peripheral blood-derived myeloid dendritic cell subsets. Front Immunol 2019;10:2440.
[6]. Cavaliere A, Ermito S, Dinatale A, et al. Management of molar pregnancy. J Prenat Med 2009;3:15–7.
[7]. Borhart J. Emergency department management of vaginal bleeding in the nonpregnant patient. Emerg Med Pract 2013;15:1–20.
[8]. Sapra KJ, Joseph KS, Galea S, et al. Signs and symptoms of early pregnancy loss. Reprod Sci 2017;24:502–13.
[9]. Redinger A, Nguyen H. Incomplete abortions. In: StatPearls. Treasure Island (FL): StatPearls Publishing 2021.
[10]. Griebel CP, Halvorsen J, Golemon TB, Day AA, et al. Management of spontaneous abortion. Am Fam Phys 2005;72:1243–50.
[11]. Gemer O, Segal S, Kopmar A, et al. The current clinical presentation of complete molar pregnancy. Arch Gynecol Obstet 2000;264:33–4.
[12]. Sebire NJ. The diagnosis of gestational trophoblastic disease in early pregnancy: implications for screening, counseling and management. Ultrasound Obstet Gynecol 2005;25:421–4.
[13]. Berkowitz RS, Goldstein DP, Bernstein MR. Natural history of partial molar pregnancy. Obstet Gynecol 1985;66:677–81.
[14]. Ross JA, Unipan A, Clarke J, et al. Ultrasound diagnosis of molar pregnancy. Ultrasound 2018;26:153–9.
[15]. Neilson JP, Gyte GM, Hickey M, et al. Medical treatments for incomplete miscarriage (less than 24 weeks). Cochrane Database Syst Rev 2010;CD007223.
[16]. Patua B, Dasgupta M, Bhattacharyya SK, et al. An approach to evaluate the efficacy of vaginal misoprostol administered for a rapid management of first trimester spontaneous onset incomplete abortion, in comparison to surgical curettage. Arch Gynecol Obstet 2013;288:1243–8.
[17]. Matteson KA, Rahn DD, Wheeler TL 2nd, et al. Nonsurgical management of heavy menstrual bleeding: a systematic review. Obstet Gynecol 2013;121:632–43.

beta-human chorionic gonadotropin; case report; dilation and curettage; incomplete abortion; mole; vaginal bleeding

Copyright © 2021 the Author(s). Published by Wolters Kluwer Health, Inc.