None, A. D. (2024). Diagnostic Performance of Pre-evacuation Ultrasound for Hydatidiform Mole in Women Undergoing Uterine Evacuation. Journal of Contemporary Clinical Practice, 10(2), 743-752.
MLA
None, Anirban Dasgupta. "Diagnostic Performance of Pre-evacuation Ultrasound for Hydatidiform Mole in Women Undergoing Uterine Evacuation." Journal of Contemporary Clinical Practice 10.2 (2024): 743-752.
Chicago
None, Anirban Dasgupta. "Diagnostic Performance of Pre-evacuation Ultrasound for Hydatidiform Mole in Women Undergoing Uterine Evacuation." Journal of Contemporary Clinical Practice 10, no. 2 (2024): 743-752.
Harvard
None, A. D. (2024) 'Diagnostic Performance of Pre-evacuation Ultrasound for Hydatidiform Mole in Women Undergoing Uterine Evacuation' Journal of Contemporary Clinical Practice 10(2), pp. 743-752.
Vancouver
Anirban Dasgupta AD. Diagnostic Performance of Pre-evacuation Ultrasound for Hydatidiform Mole in Women Undergoing Uterine Evacuation. Journal of Contemporary Clinical Practice. 2024 Jul;10(2):743-752.
Background: Hydatidiform mole may be difficult to distinguish from non-molar early pregnancy loss using ultrasound. Histopathological examination is required for confirmation.Objective:To describe histopathological diagnoses and assess the diagnostic performance of pre-evacuation ultrasound for hydatidiform mole.Methods
This prospective, observational, single-centre study included 150 women undergoing uterine evacuation for first-trimester pregnancy loss or suspected abnormal early pregnancy. Ultrasound suspicion was defined by one or more of four findings: vesicular or snowstorm pattern, absent foetal pole, uterine enlargement, or theca-lutein cysts. Histopathology was the reference standard. Follow-up outcomes were summarized descriptively.Results:Histopathology confirmed hydatidiform mole in 3 of 150 women (2.0%): 2 complete and 1 partial mole. The ultrasound screen was positive in 116 women. Sensitivity was 66.7%, specificity 22.4%, positive predictive value 1.7%, negative predictive value 97.1%, and accuracy 23.3%. One woman developed persistent gestational trophoblastic neoplasia. Among the two women without persistent disease, beta-hCG normalized at 38 and 52 days.Conclusion:The ultrasound screen had low specificity and positive predictive value. Histopathological examination remained necessary for confirmation. The small number of molar cases limited the precision of these estimates.
Keywords
Hydatidiform mole
Gestational trophoblastic disease
Ultrasonography
Histopathology
Diagnostic accuracy.
INTRODUCTION
Gestational trophoblastic disease (GTD) comprises a group of pregnancy-related disorders arising from abnormal trophoblastic proliferation. The spectrum includes hydatidiform mole, which may be complete or partial, and gestational trophoblastic neoplasia, which includes invasive mole, choriocarcinoma, placental-site trophoblastic tumour and epithelioid trophoblastic tumour. These conditions differ in their biology, behaviour and need for surveillance, making accurate classification important from the time of diagnosis.1,2
Hydatidiform mole is an abnormal conception in which the genetic contribution and development of embryonic and placental tissues differ from those of a normal pregnancy. Complete moles are usually androgenetic and lack embryonic development, whereas partial moles commonly contain an additional paternal chromosome set and may show foetal or embryonic tissue. The clinical course after evacuation also differs between these subtypes. The reported frequency of molar pregnancy varies across regions and populations, reflecting differences in ascertainment, reproductive patterns and access to diagnostic services.3,4
The clinical presentation is variable. Vaginal bleeding and symptoms associated with early pregnancy loss are common, while the classic findings of uterine enlargement, marked hyperemesis, hypertensive disease and hyperthyroidism are less consistently encountered when diagnosis occurs early. Ultrasound and quantitative serum human chorionic gonadotropin (hCG) are important in the initial assessment, but findings can overlap with non-molar miscarriage. Complete moles may produce a heterogeneous intrauterine mass with multiple small cystic spaces, while partial moles can resemble an anembryonic pregnancy or an ordinary early pregnancy failure.5,6
Histopathological examination of evacuated products of conception remains central to confirming molar disease and distinguishing it from hydropic miscarriage. Early gestational specimens may show subtle or overlapping morphology, and interpretation can vary between observers. Ancillary testing can improve classification: p57 immunohistochemistry helps identify most complete moles, while molecular genotyping can distinguish complete mole, partial mole and non-molar conceptions by assessing parental genomic contributions. These techniques are particularly useful when morphology is equivocal or when the subtype will change follow-up decisions.7–9
After uterine evacuation, serial hCG measurement is used to document regression and identify persistent disease. A plateau or rise in hCG may indicate gestational trophoblastic neoplasia and requires timely evaluation. Follow-up schedules are tailored to histological subtype and the pattern of hCG regression; reliable communication and access to monitoring are therefore integral to care. Despite advances in imaging and laboratory methods, histology and surveillance remain complementary parts of the diagnostic pathway. A clear description of how ultrasound findings are defined is also necessary when estimating diagnostic performance, because a general suspicion recorded by a clinician is not equivalent to a rule based on individual sonographic features.6,10
This manuscript examines women undergoing evacuation for early pregnancy loss or suspected abnormal early pregnancy. It describes the histopathological diagnoses, selected clinical and ultrasound characteristics, and post-evacuation follow-up.
METHODOLOGY
Study design and setting
This was a prospective, observational, single-centre diagnostic accuracy study conducted in the Department of Obstetrics and Gynaecology, Shantiniketan Medical College and Hospital, Bolpur, West Bengal, India, from February 2024 to July 2024. Consecutive eligible women presenting to the emergency department, antenatal outpatient department, ultrasound unit, and operation theatre were screened for inclusion.
The study evaluated whether pre-evacuation clinical and ultrasound findings could identify hydatidiform mole among women undergoing uterine evacuation for first-trimester pregnancy loss or suspected abnormal early pregnancy. Histopathological examination of evacuated tissue was used as the reference standard. Ultrasound is useful for raising suspicion of molar pregnancy, but histological examination remains necessary for definitive diagnosis.11,12
Ethical considerations
The study protocol was reviewed and approved by the Institutional Ethics Committee of Shantiniketan Medical College and Hospital before recruitment began. Written informed consent was obtained from all eligible participants before enrolment. Participation was voluntary, and women were free to withdraw at any time without any effect on the clinical care they received.
Patient identities were kept confidential. Each participant was assigned a serial number, and personal identifiers were not included in the study database. Data were stored in a password-protected electronic file accessible only to the principal investigator and designated research personnel. All participants received standard clinical care irrespective of study participation. Women with confirmed molar pregnancy were advised regarding serial beta-hCG surveillance, contraception during follow-up, and referral for further management if persistent gestational trophoblastic neoplasia (GTN) was suspected.11
Sample size calculation
A consecutive sample of 150 women undergoing uterine evacuation for first-trimester pregnancy loss or suspected abnormal early pregnancy was planned over the 6-month study period. The sample size was based on the expected number of eligible evacuations at the study centre during the recruitment period and the anticipated frequency of hydatidiform mole in this population.
Assuming a molar pregnancy frequency of approximately 1–3 per 1,000 pregnancies in the general population, but a higher proportion among women undergoing evacuation for abnormal early pregnancy, a target of 150 participants was considered feasible for estimating ultrasound sensitivity, specificity, positive predictive value, negative predictive value, and accuracy. Because molar pregnancy is uncommon, the number of confirmed molar cases was expected to be small; therefore, the primary analysis was descriptive and diagnostic accuracy was reported with exact 95% confidence intervals.5,12
Inclusion criteria
Women were eligible for inclusion if they fulfilled all of the following criteria:
• Age between 18 and 49 years.
• Undergoing suction evacuation or dilatation and curettage for first-trimester miscarriage, missed abortion, blighted ovum, retained products of conception, or unexplained first-trimester bleeding.
• Pre-evacuation transvaginal ultrasound and quantitative serum beta-hCG available.
• Evacuated tissue submitted for routine histopathological examination.
• Willing and able to provide written informed consent.
Exclusion criteria
Women were excluded if any of the following was present:
• Inability or unwillingness to provide informed consent.
• Confirmed ectopic pregnancy.
• Ongoing intrauterine pregnancy continuing beyond evacuation.
• Evacuated tissue not sent for histopathological examination.
• Incomplete pre-evacuation ultrasound, beta-hCG, or histopathology data.
• Known non-gestational malignancy.
Outcomes
The primary outcome was histopathologically confirmed hydatidiform mole among women undergoing uterine evacuation for first-trimester pregnancy loss or suspected abnormal early pregnancy.
The secondary outcomes were:
• Diagnostic accuracy of pre-evacuation ultrasound suspicion for hydatidiform mole.
• Differences in pre-evacuation beta-hCG, gestational age, and ultrasound findings between molar and non-molar pregnancies.
• Post-evacuation beta-hCG normalization among women with confirmed molar pregnancy.
• Persistent GTN among women with confirmed molar pregnancy during the available follow-up period.
Persistent GTN was defined as a beta-hCG plateau of four values over 3 weeks with no more than 10% change, a beta-hCG rise of more than 10% across three consecutive values over at least 2 weeks, persistent detectable beta-hCG 6 months or longer after molar evacuation, or histological confirmation of invasive mole or choriocarcinoma.6,11
Variables
The following variables were recorded for each participant:
Variable Measurement
Age Completed years
Indication for evacuation Miscarriage, missed abortion, blighted ovum, retained products, unexplained first-trimester bleeding
Gestational age Weeks, by last menstrual period and/or ultrasound
Pre-evacuation beta-hCG Quantitative serum beta-hCG in mIU/mL
Ultrasound suspicion of mole Present or absent
Vesicular/snowstorm pattern Present or absent
Absent foetal pole Present or absent
Uterine size Enlarged or normal
Theca-lutein cyst Present or absent
Histopathology Molar or non-molar; complete or partial mole if confirmed
Persistent GTN Present or absent during available follow-up
Time to beta-hCG normalization Number of days from evacuation to the first normal β-hCG result. For complete mole, normalization was confirmed by three consecutive normal weekly values. For partial mole, the time recorded was the first normal result; confirmation followed the study’s partial-mole follow-up schedule.
Procedures
Eligible women were identified consecutively at the time of presentation for uterine evacuation. After informed consent, demographic details, obstetric history, clinical symptoms, and indication for evacuation were recorded using a structured case-report form.
A quantitative serum beta-hCG assay was performed before evacuation using the hospital’s routine electrochemiluminescence immunoassay platform. Transvaginal pelvic ultrasonography was performed by a qualified obstetrician or radiologist using a conventional 2D ultrasound machine. The ultrasound report was reviewed for the presence of a vesicular or snowstorm pattern, absent foetal pole, uterine enlargement, and theca-lutein cysts. Ultrasound suspicion of molar pregnancy was defined as the presence of one or more of these findings. For analysis, this composite variable was recalculated from the four recorded component findings because the pre-existing composite flag did not match this definition in 103 records.
Uterine evacuation was performed according to routine departmental practice. Suction evacuation, with or without gentle curettage, was the preferred method for evacuation of suspected molar pregnancy. Oxytocin infusion was used during or after evacuation when clinically indicated to reduce uterine bleeding. All evacuated tissue was sent for histopathological examination. The final diagnosis was based on histopathology, and molar pregnancies were classified as complete or partial where possible.6,11
Techniques
Pre-evacuation ultrasound was performed using a transvaginal 5–9 MHz endocavitary probe. Standard pelvic ultrasound views were obtained to assess uterine size, endometrial cavity contents, presence or absence of foetal pole, adnexal masses, and theca-lutein cysts. Colour Doppler was used only when clinically indicated and was not a mandatory study variable. A vesicular or snowstorm pattern, absent foetal pole, enlarged uterus, and theca-lutein cysts were recorded as binary findings.5,12
Quantitative serum beta-hCG was measured using a conventional electrochemiluminescence immunoassay available in the hospital laboratory. Blood samples were collected before evacuation and processed. Beta-hCG values were recorded in mIU/mL.
Histopathological examination was performed on 10% buffered formalin-fixed, paraffin-embedded tissue sections stained with haematoxylin and eosin. The pathologist assessed chorionic villous morphology, trophoblastic proliferation, hydropic change, and the presence or absence of foetal tissue. Complete and partial moles were classified according to routine histopathological criteria. Immunohistochemistry, including p57 staining, and molecular genetic testing were not routinely performed and were not part of the study protocol.6,11
Follow-up
Women with histopathologically confirmed molar pregnancy were advised to undergo serial quantitative serum beta-hCG monitoring. Beta-hCG was measured weekly until three consecutive normal values were obtained, followed by monthly monitoring for 6 months in women with complete molar pregnancy. Women with partial molar pregnancy were advised to have a confirmatory beta-hCG test 1 month after the first normal value.6,11
Persistent GTN was suspected if beta-hCG failed to decline appropriately, plateaued, or rose during follow-up. Women with suspected persistent GTN were assessed clinically and with pelvic ultrasound, and were referred for further management according to institutional protocol. Chemotherapy was not routinely administered after molar evacuation unless persistent GTN was confirmed.
Statistical analysis
Data were entered in Microsoft Excel and analysed using standard statistical software. Categorical variables were summarized as frequency and percentage. Continuous variables were summarized as mean ± standard deviation or median with interquartile range, depending on distribution.
The frequency of histopathologically confirmed hydatidiform mole was reported as a proportion with an exact 95% confidence interval. Diagnostic accuracy of ultrasound was reported as sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy. Chi-square test or Fisher’s exact test was used to compare categorical variables between molar and non-molar groups. Mann–Whitney U test was used to compare beta-hCG and gestational age between groups. No regression analysis, risk-score development, or multivariable modelling was performed.
RESULTS
A total of 150 women undergoing uterine evacuation for first-trimester pregnancy loss or suspected abnormal early pregnancy were included. The mean maternal age was 27.4 ± 5.6 years, and the median gestational age at evacuation was 8.5 weeks (interquartile range [IQR] 7.3–10.2). The median pre-evacuation serum beta-human chorionic gonadotropin (beta-hCG) level was 11,837 mIU/mL (IQR 7,013–23,949). Missed abortion was the most common indication for evacuation, accounting for 52 women (34.7%), followed by blighted ovum in 32 women (21.3%) and incomplete miscarriage in 31 women (20.7%). Table 1 summarizes the baseline characteristics and indications for uterine evacuation.
Table 1. Baseline characteristics of study participants (N = 150)
Variable Mean ± SD Median (IQR)
Age, years 27.4 ± 5.6 27 (23–32)
Gestational age, weeks 8.8 ± 2.0 8.5 (7.3–10.2)
Pre-evacuation beta-hCG, mIU/mL 17,480 ± 13,260 11,837 (7,013–23,949)
Indication for evacuation n %
Missed abortion 52 34.7
Blighted ovum 32 21.3
Incomplete miscarriage 31 20.7
Retained products of conception 18 12.0
Unexplained first-trimester bleeding 17 11.3
Histopathological examination confirmed hydatidiform mole in 3 of 150 women, giving an overall frequency of 2.0% (95% confidence interval [CI] 0.2–5.7%). Complete hydatidiform mole was identified in 2 women (1.3%), whereas partial hydatidiform mole was identified in 1 woman (0.7%). The remaining 147 women (98.0%) had non-molar histopathology. Table 2 presents the final histopathological diagnoses and the frequency of molar disease.
Table 2. Final histopathological diagnosis and frequency of hydatidiform mole (N = 150)
Final histopathological diagnosis n %
Non-molar 147 98.0
Complete hydatidiform mole 2 1.3
Partial hydatidiform mole 1 0.7
Total 150 100.0
Measure n/N Percentage (95% CI)
Histopathologically confirmed hydatidiform mole 3/150 2.0 (0.2–5.7)
There was no significant difference in maternal age or gestational age between women with molar and non-molar pregnancies. The median age was 24.0 years in the molar group and 27.0 years in the non-molar group (Mann–Whitney U = 150.5; p = 0.350), while the median gestational age was 7.9 weeks and 8.5 weeks, respectively (Mann–Whitney U = 195.0; p = 0.737). In contrast, pre-evacuation beta-hCG was significantly higher among women with molar pregnancy, with a median of 42,000 mIU/mL compared with 11,565 mIU/mL in non-molar pregnancies (Mann–Whitney U = 394.0; p = 0.020). Table 3 compares the continuous variables between the two groups.
Table 3. Comparison of maternal age, gestational age, and beta-hCG between molar and non-molar pregnancies
Variable Molar (n = 3), median (IQR) Non-molar (n = 147), median (IQR) Mann–Whitney U p value
Age, years 24.0 (22.5–26.0) 27.0 (23.0–32.0) 150.5 0.350
Gestational age, weeks 7.9 (7.2–9.4) 8.5 (7.3–10.1) 195.0 0.737
Pre-evacuation beta-hCG, mIU/mL 42,000 (36,500–43,500) 11,565 (6,838–22,920) 394.0 0.020
A vesicular or “snowstorm” ultrasound pattern was not significantly associated with histopathologically confirmed molar pregnancy in this cohort. The pattern was present in 1 of 3 molar pregnancies and 11 of 147 non-molar pregnancies (Fisher’s exact test; p = 0.223). Theca-lutein cysts were significantly associated with molar pregnancy, being present in 1 of 3 molar cases and 1 of 147 non-molar cases (Fisher’s exact test; p = 0.040). Absent foetal pole and uterine enlargement were not significantly associated with molar pregnancy. Table 4 presents the association between ultrasound findings and final histopathological diagnosis.
Table 4. Association between ultrasound findings and histopathologically confirmed hydatidiform mole
Ultrasound finding Category Non-molar (n = 147) Molar (n = 3) p value
Vesicular/snowstorm pattern Yes 11 1 0.223
Vesicular/snowstorm pattern No 136 2 0.223
Absent foetal pole Yes 110 2 1.000
Absent foetal pole No 37 1 1.000
Uterine size Enlarged 29 2 0.109
Uterine size Normal 118 1 0.109
Theca-lutein cyst Yes 1 1 0.040
Theca-lutein cyst No 146 2 0.040
Using the protocol-defined composite of one or more ultrasound findings, 2 of 3 molar pregnancies screened positive and 114 of 147 non-molar pregnancies also screened positive. Sensitivity was 66.7% (95% CI 9.4–99.2%), specificity 22.4% (95% CI 16.0–30.1%), positive predictive value 1.7% (95% CI 0.2–6.1%), negative predictive value 97.1% (95% CI 84.7–99.9%), and overall accuracy 23.3% (95% CI 16.8–30.9%). Table 5 presents the recalculated diagnostic performance.
Table 5. Diagnostic accuracy of pre-evacuation ultrasound for hydatidiform mole
Measure Value
True positive 2
False positive 114
False negative 1
True negative 33
Sensitivity 66.7% (95% CI 9.4–99.2%)
Specificity 22.4% (95% CI 16.0–30.1%)
Positive predictive value 1.7% (95% CI 0.2–6.1%)
Negative predictive value 97.1% (95% CI 84.7–99.9%)
Accuracy 23.3% (95% CI 16.8–30.9%)
Among the 3 women with histopathologically confirmed molar pregnancy, 2 had complete hydatidiform mole and 1 had partial hydatidiform mole. Ultrasound suspicion was present in 2 of the 3 molar pregnancies; the partial molar pregnancy was not suspected on pre-evacuation ultrasound. A vesicular/snowstorm pattern was present in only 1 molar pregnancy, while theca-lutein cysts were present in 1 molar pregnancy and 1 non-molar pregnancy. One woman with complete hydatidiform mole developed persistent GTN during the available follow-up period. Table 6 presents the clinical and ultrasound characteristics of the confirmed molar pregnancies.
Table 6. Characteristics of women with histopathologically confirmed hydatidiform mole
Sl. No. Age (years) Gestational age (weeks) Pre-evacuation beta-hCG (mIU/mL) Ultrasound suspicion Vesicular/snowstorm pattern Theca-lutein cyst Histopathology (hydatidiform mole) Persistent GTN Time to beta-hCG normalization (days)
73 21 7.9 42,000 Yes Yes Yes Complete No 38
96 24 6.5 45,000 Yes No No Complete Yes Not applicable
143 28 10.9 31,000 No No No Partial No 52
Among the 2 women with molar pregnancy who did not develop persistent GTN, beta-hCG normalized in both cases, at 38 and 52 days after evacuation. One of the 3 women with confirmed molar pregnancy developed persistent GTN, giving a persistent GTN proportion of 33.3% (95% CI 0.6–90.6%). Because only one GTN event occurred among three molar cases, this estimate was interpreted as descriptive rather than as a precise measure of local risk. Table 7 summarizes post-evacuation beta-hCG follow-up and persistent GTN among confirmed molar cases.
Table 7. Post-evacuation beta-hCG follow-up and persistent GTN among confirmed molar pregnancies (n = 3)
Outcome n/N Percentage (95% CI) Median days to beta-hCG normalization
Beta-hCG normalization among all molar cases 2/3 66.7% (9.4–99.2%) 45
Persistent GTN among confirmed molar cases 1/3 33.3% (0.6–90.6%) Not applicable
Overall, pre-evacuation beta-hCG and theca-lutein cysts were associated with histopathologically confirmed hydatidiform mole. The protocol-defined ultrasound screen had moderate sensitivity but low specificity and positive predictive value; histopathological confirmation was necessary for definitive classification. The small number of confirmed molar cases and the single persistent GTN event limited the precision of the follow-up estimates. Table 8 summarizes the principal study findings.
Table 8. Summary of principal findings
Finding Result
Total participants 150
Histopathologically confirmed hydatidiform mole 3/150
(2.0%; 95% CI 0.2–5.7%)
Complete hydatidiform mole 2/150 (1.3%)
Partial hydatidiform mole 1/150 (0.7%)
Ultrasound suspicion of molar pregnancy 116/150 (77.3%)
Ultrasound sensitivity 66.7% (95% CI 9.4–99.2%)
Ultrasound specificity 22.4% (95% CI 16.0–30.1%)
Ultrasound positive predictive value 1.7% (95% CI 0.2–6.1%)
Persistent GTN among molar cases 1/3 (33.3%; 95% CI 0.6–90.6%)
Median time to beta-hCG normalization 45 days among non-persistent molar cases
Ultrasound negative predictive value 97.1% (95% CI 84.7–99.9%)
Ultrasound accuracy 23.3% (95% CI 16.8–30.9%)
DISCUSSION
These findings reinforce the difficulty of identifying hydatidiform mole before evacuation in patients presenting with early pregnancy loss. Ultrasound performance varies across studies and depends on case selection and how a positive scan is defined. Fowler et al. (2006)12 reported that routine ultrasound detected complete moles more often than partial moles, with overall sensitivity of 44% and specificity of 74%. Kirk et al. (2007)13 also found lower sensitivity for partial than complete mole; their study included women with suspected or histologically confirmed disease, so its estimates reflect that selected diagnostic pathway. In a missed-miscarriage cohort, Johns et al. (2005)14 reported a 56% detection rate using ultrasound alone.
Results from larger or differently defined cohorts also vary. Ross et al. (2018)15 reported ultrasound sensitivity and specificity for gestational trophoblastic disease overall, while separately finding greater preoperative detection for complete than partial mole. Because the overall estimates were for GTD, they are not a direct comparison with our histopathology-based molar-pregnancy analysis. Memtsa et al. (2020),16 studying women with missed miscarriage, likewise found higher ultrasound detection for complete than partial mole and substantial variation across published cohorts. Newhouse et al. (2022)17 pooled studies across pregnancy trimesters and reported moderate sensitivity with high specificity. Differences in populations, gestational age, referral setting, reader expertise and scan definitions likely contribute to this variation.
Our operational rule treated any one of four recorded features as a positive screen. That is different from an overall clinician or sonographer impression of suspected mole, as used in several comparative studies. The original composite suspicion flag also differed from the feature-based rule in many records. Recalculation made the diagnostic table consistent with the Methods, but
does not establish that this broad rule is clinically valid. The distinction matters because components such as absent foetal pole can occur in non-molar early pregnancy loss. Jauniaux et al. (2020)18 found only fair-to-moderate agreement between initial and secondary ultrasound assessments, and Ross et al. (2018)15 noted that diagnostic criteria remained ill-defined.
The beta-hCG findings are consistent with its role as an adjunct to imaging, rather than a stand-alone diagnostic test. Johns et al. (2005)14 found elevated values in many of the cases for which measurements were available. Memtsa et al. (2020),16 however, reported both low and high values across complete and partial moles, illustrating that a single threshold may not identify all cases.
Histopathology remains important because ultrasound and morphology can both have limitations. Sebire et al. (2001)19 found that many histologically confirmed moles had been interpreted sonographically as missed miscarriage or anembryonic pregnancy, particularly partial moles. In the present study, histopathology provides the reference classification, but the manuscript states that p57 staining and molecular testing were not performed. Ronnett et al. (2011)8 describe how p57 can help identify complete mole, while genotyping can distinguish complete mole, partial mole and non-molar specimens. Vang et al. (2012)9 found that adding p57 improved complete-mole diagnosis, but did not resolve the distinction between partial mole and non-molar specimens. Heller (2018)7 also characterizes GTD as a challenging area for pathological evaluation.
The follow-up findings should be interpreted separately from studies that specifically measured GTN risk after hCG normalization. The present study records a persistent GTN event, but it does not provide a comparable post-normalization risk estimate. Schmitt et al. (2013)20 reported that post-normalization GTN was uncommon and varied by mole type. Coyle et al. (2017)21 found a higher post-normalization risk for complete than partial mole, with risk also related to the time taken for hCG to normalize. Albright et al. (2020)22 reached a similar conclusion in a systematic review and meta-analysis. Their larger follow-up populations and defined post-normalization endpoint differ substantially, so their estimates should contextualize surveillance rather than serve as direct comparators.
Overall, the literature and this analysis support using ultrasound to raise suspicion, histopathology to confirm and classify molar disease, and serial hCG measurements to monitor regression.
Strengths and Limitations
The study includes histopathological diagnoses and records several clinical, ultrasound and follow-up variables, allowing the diagnostic rule to be examined transparently. Recalculating the composite screen from its stated components also exposed its difference from the original suspicion flag. The small number of molar cases makes subgroup and follow-up estimates highly imprecise, and the feature-based screen has not been clinically validated.
CONCLUSION
In this study, histopathology confirmed three molar pregnancies among 150 women undergoing uterine evacuation. The protocol-defined ultrasound screen detected two cases but also yielded numerous false-positive results, resulting in low specificity and positive predictive value. One woman with a molar pregnancy developed persistent gestational trophoblastic neoplasia; beta-hCG normalized in the other two cases. These findings support the role of ultrasound in identifying features that may raise suspicion, while histopathological examination remains necessary to confirm molar disease. The small number of confirmed cases and the single persistent GTN event limited the precision and interpretation of the diagnostic and follow-up estimates.
REFERENCES
1. Altieri A, Franceschi S, Ferlay J, Smith J, La Vecchia C. Epidemiology and aetiology of gestational trophoblastic diseases. Lancet Oncol. 2003;4(11):670-8. doi:10.1016/S1470-2045(03)01245-2.
2. Seckl MJ, Sebire NJ, Berkowitz RS. Gestational trophoblastic disease. Lancet. 2010;376(9742):717-29. doi:10.1016/S0140-6736(10)60280-2.
3. Berkowitz RS, Goldstein DP. Clinical practice: molar pregnancy. N Engl J Med. 2009;360(16):1639-45. doi:10.1056/NEJMcp0900696.
4. Lurain JR. Gestational trophoblastic disease I: epidemiology, pathology, clinical presentation and diagnosis of gestational trophoblastic disease, and management of hydatidiform mole. Am J Obstet Gynecol. 2010;203(6):531-9. doi:10.1016/j.ajog.2010.06.073.
5. Shaaban AM, Rezvani M, Haroun RR, Kennedy AM, Elsayes KM, Olpin JD, et al. Gestational trophoblastic disease: clinical and imaging features. Radiographics. 2017;37(2):681-700. doi:10.1148/rg.2017160140.
6. Ngan HYS, Seckl MJ, Berkowitz RS, Xiang Y, Golfier F, Sekharan PK, et al. Diagnosis and management of gestational trophoblastic disease: 2021 update. Int J Gynaecol Obstet. 2021;155 Suppl 1:86-93. doi:10.1002/ijgo.13877.
7. Heller DS. Update on the pathology of gestational trophoblastic disease. APMIS. 2018;126(7):647-54. doi:10.1111/apm.12786.
8. Ronnett BM, DeScipio C, Murphy KM. Hydatidiform moles: ancillary techniques to refine diagnosis. Int J Gynecol Pathol. 2011;30(2):101-16. doi:10.1097/PGP.0b013e3181f4de77.
9. Vang R, Gupta M, Wu LS, Yemelyanova AV, Kurman RJ, Murphy KM, et al. Diagnostic reproducibility of hydatidiform moles: ancillary techniques (p57 immunohistochemistry and molecular genotyping) improve morphologic diagnosis. Am J Surg Pathol. 2012;36(3):443-53. doi:10.1097/PAS.0b013e31823b13fe.
10. Joyce CM, Fitzgerald B, McCarthy TV, Coulter J, O'Donoghue K. Advances in the diagnosis and early management of gestational trophoblastic disease. BMJ Med. 2022;1(1):e000321. doi:10.1136/bmjmed-2022-000321.
11. Tidy J, Seckl M, Hancock BW, on behalf of the Royal College of Obstetricians and Gynaecologists. Management of gestational trophoblastic disease. BJOG. 2021;128:e1-27. doi:10.1111/1471-0528.16266.
12. Fowler DJ, Lindsay I, Seckl MJ, Sebire NJ. Routine pre-evacuation ultrasound diagnosis of hydatidiform mole: experience of more than 1000 cases from a regional referral center. Ultrasound Obstet Gynecol. 2006;27(1):56-60. doi:10.1002/uog.2592.
13. Kirk E, Papageorghiou AT, Condous G, Bottomley C, Bourne T. The accuracy of first trimester ultrasound in the diagnosis of hydatidiform mole. Ultrasound Obstet Gynecol. 2007;29(1):70-5. doi:10.1002/uog.3875.
14. Johns J, Greenwold N, Buckley S, Jauniaux E. A prospective study of ultrasound screening for molar pregnancies in missed miscarriages. Ultrasound Obstet Gynecol. 2005;25(5):493-7. doi:10.1002/uog.1888.
15. Ross JA, Unipan A, Clarke J, Magee C, Johns J. Ultrasound diagnosis of molar pregnancy. Ultrasound. 2018;26(3):153-9. doi:10.1177/1742271X17748514.
16. Memtsa M, Johns J, Jurkovic D, Ross JA, Sebire NJ, Jauniaux E. Diagnosis and outcome of hydatidiform moles in missed miscarriage: a cohort study, systematic review and meta-analysis. Eur J Obstet Gynecol Reprod Biol. 2020;253:206-12. doi:10.1016/j.ejogrb.2020.07.030.
17. Newhouse I, Spacey A, Scragg B, Szczepura K. The diagnostic value and accuracy of ultrasound in diagnosing hydatidiform mole: a systematic review and meta-analysis of the literature. Radiography (Lond). 2022;28(4):897-905. doi:10.1016/j.radi.2022.06.005.
18. Jauniaux E, Memtsa M, Johns J, Ross JA, Sebire NJ, Jurkovic D. Ultrasound diagnosis of complete and partial hydatidiform moles in early pregnancy failure: an inter-observer study. Placenta. 2020;97:65-7. doi:10.1016/j.placenta.2020.06.013.
19. Sebire NJ, Rees H, Paradinas F, Seckl M, Newlands E. The diagnostic implications of routine ultrasound examination in histologically confirmed early molar pregnancies. Ultrasound Obstet Gynecol. 2001;18(6):662-5. doi:10.1046/j.0960-7692.2001.00589.x.
20. Schmitt C, Doret M, Massardier J, Hajri T, Schott AM, Raudrant D, et al. Risk of gestational trophoblastic neoplasia after hCG normalisation according to hydatidiform mole type. Gynecol Oncol. 2013;130(1):86-9. doi:10.1016/j.ygyno.2013.03.010.
21. Coyle C, Short D, Jackson L, Sebire NJ, Kaur B, Harvey R, et al. What is the optimal duration of human chorionic gonadotrophin surveillance following evacuation of a molar pregnancy? A retrospective analysis on over 20,000 consecutive patients. Gynecol Oncol. 2017;148(2):254-7. doi:10.1016/j.ygyno.2017.12.008.
22. Albright BB, Shorter JM, Mastroyannis SA, Ko EM, Schreiber CA, Sonalkar S. Gestational trophoblastic neoplasia after human chorionic gonadotropin normalization following molar pregnancy: a systematic review and meta-analysis. Obstet Gynecol. 2020;135(1):12-23. doi:10.1097/AOG.0000000000003566.
Recommended Articles
Original Article
Effects of Large-Volume Infusion of Ringer’s Lactate Versus 0.9% Saline on Acid–Base Balance and Serum Electrolytes During Spine Surgery: A Randomised Comparative Study