|Year : 2021 | Volume
| Issue : 3 | Page : 281-287
Epidemiological aspects and diagnostic accuracy of morphological diagnosis of hydatidiform mole using p57kip2 immunostain in Nnewi, South-East Nigeria – A multicenter study
Chinedu Onwuka Ndukwe, Cornelius Ozobia Ukah
Department of Anatomic Pathology and Forensic Medicine, Nnamdi Azikiwe University, Awka, Anambra State, Nigeria
|Date of Submission||24-Dec-2020|
|Date of Decision||04-Jan-2021|
|Date of Acceptance||01-Feb-2021|
|Date of Web Publication||26-Jul-2021|
Chinedu Onwuka Ndukwe
Department of Anatomic Pathology and Forensic Medicine, Nnamdi Azikiwe University, Awka, Nnewi Campus, Anambra State
Source of Support: None, Conflict of Interest: None
Introduction: There have been numerous studies on hydatidiform mole (HM) in Nigeria, but some lacked histological diagnosis, and others did not differentiate complete hydatidiform mole (CHM) from partial hydatidiform mole (PHM). In addition, none of these studies involved immunohistochemical (IHC) diagnosis or assessed the accuracy of morphologic diagnosis of CHM. The objective of this study is to determine the epidemiology and accuracy of morphologic diagnosis of CHM using p57KIP2 IHC stain. Materials and Methods: The paraffin blocks of histologically diagnosed cases of CHM and PHM seen at two histopathology laboratories in Nnewi, South-East Nigeria, were retrieved from the archives. IHC staining for p57KIP2 was done. Results: In this study, we reviewed and reclassified 54 cases of routinely stained HMs as 15 (27.8%) CHMs and 39 (72.2%) PHMs. However, following IHC staining, we further reclassified them as 21 (38.9%) CHMs and 33 (61.1%) PHMs. Discordant histopathological diagnosis between routine staining with hematoxylin and eosin (H and E) and IHC staining with p57KIP2 was identified in eight cases (which constitutes 14.8% of the studied cases): one case was a false-positive diagnosis, while the remaining seven cases were false-negative diagnosis. Thus, the sensitivity and specificity of histopathological diagnosed cases of CHM by H and E were 66.7% and 97%, respectively, with a positive predictive value of 93.3%, negative predictive value of 82.1%, and total accuracy of 85.2%. Conclusions: Morphology alone is suboptimal for accurate diagnosis of CHM. We therefore strongly recommend the routine use of p57KIP2 immunostain in all suspected cases of HM.
Keywords: Complete hydatidiform mole, diagnostic accuracy, immunohistochemistry, Nigeria, p57KIP2
|How to cite this article:|
Ndukwe CO, Ukah CO. Epidemiological aspects and diagnostic accuracy of morphological diagnosis of hydatidiform mole using p57kip2 immunostain in Nnewi, South-East Nigeria – A multicenter study. J Nat Sci Med 2021;4:281-7
|How to cite this URL:|
Ndukwe CO, Ukah CO. Epidemiological aspects and diagnostic accuracy of morphological diagnosis of hydatidiform mole using p57kip2 immunostain in Nnewi, South-East Nigeria – A multicenter study. J Nat Sci Med [serial online] 2021 [cited 2023 Jan 28];4:281-7. Available from: https://www.jnsmonline.org/text.asp?2021/4/3/281/322325
| Introduction|| |
Hydatidiform mole (HM) is an abnormal conception characterized by enlarged, edematous, and vesicular chorionic villi accompanied by villous trophoblastic hyperplasia. HM is subdivided into complete hydatidiform mole (CHM) and partial hydatidiform mole (PHM) based on morphologic, cytogenetic, and clinicopathological features.
Accurate subclassification of molar specimens into CHM and PHM and distinction of these from nonmolar specimens are fundamental to clinical management and accurate assessment of the risk of persistent gestational trophoblastic disease (GTD) (invasive mole) or choriocarcinoma. The risk of persistent GTD or choriocarcinoma for CHMs (15%–20%) is significantly higher than for PHMs (0.2%–4%). Furthermore, there is the possibility of anticipating the discharge of patients with partial mole from the remission of the disease. This represents lower costs for postmolar follow-up and work absenteeism, in addition to improving the psychic impact and anticipating the reproductive return of these patients.
Cytogenetic analysis is costly, labor intensive, and is not available in most pathology laboratories. A useful marker for the differential diagnosis of CHM is p57KIP2 protein, a cell cycle inhibitor and tumor suppressor protein encoded by TP57 gene that is imprinted in the paternal chromosome. It is well expressed in the cytotrophoblast and villous mesenchyme of normal pregnancy, spontaneous abortuses, and PHMs but absent or markedly decreased in CHMs because both copies of the TP57 gene are of paternal origin.
The ratio of CHM versus PHMs varies among studies. There have been numerous studies in Nigeria, but many of them were hospital-based data, some lacked histological diagnosis, and others did not differentiate CHM from PHM. In addition, none of these studies involved immunohistochemical (IHC) diagnosis or assessed the diagnostic accuracy of morphology alone in diagnosis of CHM. All these aforementioned observations prompted this study.
The objective of this study is to determine the epidemiology and diagnostic accuracy of morphology alone in the diagnosis of CHM using p57KIP2 immunohistochemistry.
| Materials and Methods|| |
This study involves the two histopathology laboratories in Nnewi, namely Histopathology Department of Nnamdi Azikiwe University Teaching Hospital (NAUTH) – a tertiary hospital-based laboratory, and Pathocon Specialist Clinic and Research Institute (PATHOCON) – an independent histopathology laboratory.
All the formalin-fixed paraffin-embedded tissue blocks of histologically diagnosed cases of CHM and PHM in both facilities from January 2010 to December 2019 were retrieved from the archives. The cases had been diagnosed using only hematoxylin and eosin (H and E) stains.
First, the H and E-stained slides of the retrieved tissue blocks were reviewed for morphological consistency by the authors. These cases were reviewed and reclassified using six histological features: two populations of chorionic villi, central cistern, circumferential trophoblastic hyperplasia, atypia of intermediate trophoblast, scalloped outline of chorionic villi, and inclusion of trophoblast within the chorionic villous stroma.
Then sections were made from these blocks and stained with p57 immunostain by the indirect immunoperoxidase method using Thermo Fisher p57kip2 clone 57P06 at a dilution of 1:100.
The p57 immunostain was interpreted as “negative” and satisfactory when maternal decidua and/or intermediate trophoblastic cells exhibit nuclear expression of p57 (serving as internal positive control in all the cases, including CHMs), but villous stromal cells and cytotrophoblast are either entirely negative or demonstrate only limited expression (nuclear staining in <10% of these cell types). This negative result was then interpreted as consistent with a diagnosis of a CHM, provided that the morphology of the specimen was appropriate.
Specimens were interpreted as “positive” for p57 staining when there is distinct nuclear staining of villous stromal cells and cytotrophoblasts in more than 10% of these cell types.
We used p57kip2 as a reference to determine the accuracy of morphologic (H and E) diagnosis. While CHMs are the positive cases, PHMs were included as negative cases.
The minimum sample size of HM was determined using the formula, n = Z2pq/d2 where:
n = Minimum sample size
z = Percentage point of standard normal distribution curve, which corresponds to 95% confidence interval z = 1.96
p = Prevalence rate of HM in Nnewi 0.4%
q = Complimentary probability, q = 0.996
d = Degree of precision at 95% confidence limit, d = 2% = 0.02
n = (1.96)2 × 0.004 × 0.996/0.0004 = 39.
An attrition rate of 20% was assumed to account for anticipated dropout. This brought the total sample size to = calculated sample size (39)/1 − attrition rate = 39/1 − 0.2 = 39/0.8 = 48.75. This we approximated to 50.
Data were analyzed using simple descriptive statistics such as sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and Chi-square test to measure associations with level of significance pegged at P < 0.05. The above analysis was done using the Statistical Package for Social Sciences 20.0, IBM, Chicago, IL, USA.
Ethical approval for the study was obtained from the Ethics Committee of NAUTH, Nnewi.
| Results|| |
There were 506 products of conception (POCs) in the laboratory records, made up of 353 POCs and 153 POCs from NAUTH and PATHOCON laboratories, respectively. Overall, these consisted of 121 cases of products of tubal gestation and 385 products of intrauterine gestation. Of all these, 88 (17.4%) were GTDs: 70 cases of HM and 18 cases choriocarcinoma. The 70 cases of HM were derived as follows: 37 cases from NAUTH and 33 cases from PATHOCON. As is evident from the foregoing, HM constituted 79.5% of GTDs and 13.8% of POCs. For the individual laboratories, HM constituted 10.5% of POCs in NAUTH and 21.6% of POCs in PATHOCON.
We excluded 16 cases of HMs from this study due to loss of paraffin tissue blocks, lack of adequate clinical data on request forms, or poor tissue block condition leading to inability to retrieve antigen during immunohistochemistry. Fifty-four cases of HM were therefore enrolled in this study. Among these, 16 (29.6%) were CHMs while 38 (70.4%) were PHMs.
Following review by the authors, three cases (5.6%) were reclassified as follows: two cases of CHM were reclassified as PHM, while one case of PHM was reclassified as CHM. Hence, the final consensus diagnosis on H and E staining consisted of 15 (27.8%) CHMs and 39 (72.2%) PHMs. The inter-rater agreement between the original diagnosis and the reviewed diagnosis was 94.4% for concordance rate and 0.86 (very good) for Cohen's kappa.
Following IHC staining for p57KIP2, the cases were further reclassified as CHM 21 (38.9%) and PHM 33 (61.1%) giving the ratio of CHM to PHM as 1:1.6 [Figure 1].
|Figure 1: Upper panel: Complete mole with positive staining of intermediate trophoblast but no reactivity in cytotrophoblasts and villous stromal cells (p57 immunohistochemistry, ×100). Lower panel: Partial mole with positive staining in cytotrophoblast and villous stromal cells (p57 immunohistochemistry, ×200)|
Click here to view
The ages ranged from 20 to 51 years for CHM and from 15 to 51 years for PHM. The peak incidence (57.4%) falls within the ages of 25–34 years for both types of HM. Thereafter, the incidence of both CHM and PHM dropped consistently from 35 years to 49 years of age before rising slightly again [Figure 2].
|Figure 2: Frequency bar chart of the different age groups of patients with hydatidiform mole|
Click here to view
The mean age for all cases of HM was 32.1 ± 7.7 years. The mean age for CHM was slightly but insignificantly higher than that for PHM. They were 32.8 ± 8.4 years and 31.6 ± 7.3 years, respectively [Table 1].
|Table 1: Comparison of complete hydatidiform mole and partial hydatidiform mole with respect to age of patients, gestational age, and sites of involvement|
Click here to view
The majority (83.3%) of the HMs occurred in women under 40 years. Most (66.7%) of the HMs that occurred in women <40 years were PHM. In contrast, the majority (66.7%) of the HMs that occurred in women 40 years and above were CHM. Comparing the percentages of CHM within the early (<25 years), mid (25–39 years), and late (≥40 years) reproductive age groups, we found that CHM was more frequent at both extremes of maternal age (42.9% and 66.7%) than the mid-reproductive age group (31.6%) [Table 1].
Gestational age distribution
As shown in [Table 1], the mean gestational ages for all cases of HM, CHM, and PHM are 13.2 ± 3.4 weeks, 12.3 ± 3.2 weeks, and 14.7 ± 3.3 weeks, respectively. The gestational ages ranged from 6 to 22 weeks for CHM and 9–22 weeks for PHM. The gestational age modal classes for all cases of HM and CHM are both in the first trimester of pregnancy (≤13 weeks) with 50% and 66.7% of the cases, respectively, while it is the early second trimester (14–20 weeks) with 52.4% for PHM.
Distribution according to site
With respect to sites of involvement, [Table 1] shows that seven cases of HM were products of ectopic (tubal) gestation comprising 5.8% of all ectopic gestations. However, there was no case of ectopic (tubal) CHM.
Two populations of villi were present in 52.4% and 87.9% of CHMs and PHMs, respectively. Cisterns were present in 76.2% of CHMs but only 21.2% of PHMs. There was circumferential trophoblastic hyperplasia in 52.4% of CHMs but only 3% of PHMs. Furthermore, 57.1% of CHMs versus 3% of PHMs showed trophoblastic atypia. Therefore, in this study, two populations of villi, central cistern, circumferential trophoblastic hyperplasia, and atypia of intermediate trophoblast are significant (P < 0.05) features for differentiating CHM from PHM. Scalloped outline and trophoblastic inclusion are insignificant features for differentiating CHM from PHM. Central cistern, circumferential trophoblastic hyperplasia, and atypia of intermediate trophoblast are present in CHM more often than in PHM. Whereas, two populations of villi are present more in PHM than CHM [Table 2].
|Table 1: Comparison of complete hydatidiform mole and partial hydatidiform mole with respect to age of patients,|
gestational age, and sites of involvement
Click here to view
Discordant diagnosis between histopathology and IHC was identified in eight cases (14.8%); one case of CHM exhibited positive p57KIP2 and seven cases of PHM displayed negative p57KIP2. The sensitivity and specificity of histological diagnosis of CHM by H and E are 66.7% and 97%, respectively, PPV of 93.3%, NPV of 82.1%, and accuracy of 85.2% [Table 3].
| Discussion|| |
HM constituted 13.8% of POCs in this study. This finding is obviously higher than those of two previous studies from Nigeria that reported that HM formed 7.5% and 8.4% of all POCs, respectively., This apparently higher figure might be explained by the fact that the two previous studies were restricted to a tertiary hospital setting in which all POCs are more likely to be routinely sent for histological evaluation, whereas this study included an independent laboratory with more patronage of general practitioners. It is common knowledge that this group of practitioners does not routinely send every POC for histological evaluation except there is a clinical suspicion of GTD. Hence, there is a higher incidence of HM among the pool of POCs in the independent laboratory (21.6%) compared to the teaching hospital laboratory (10.5%).
In this study, CHM constitutes 38.9% of HMs which is within the range of 28.2%–60.7% reported in other studies within Nigeria.,,,, In general, the incidence of HM, especially CHM, is higher in Asia than other regions of the world.
The peak age incidence for all cases of HM in this study was 25–34 years (57.4%), while the mean age for all cases of HM, CHM, and PHM was 32.1 ± 7.7 years, 32.8 ± 8.4 years, and 31.6 ± 7.3 years, respectively. These agree with an earlier published study from Nnewi and other parts of the world.,,,, However, results from the northern part of the country were generally lower than these as the mean age of all cases of HM is 25.7 years in Zaria and 26.5 years in Gombe., These may be explained by the generally younger ages of marriage among women in the northern parts of the country and hence younger childbearing ages compared with the south-eastern part of the country.
The study by Savage et al. provides detailed data regarding the risk of PHM and CHM with increasing maternal age and confirms that the risk of PHM varies relatively little with age while CHM contributes to the main component of the overall increase with age.
In this study, nine cases (16.7%) of HM occurred after age 40 years of which six (66.7%) were CHM. In addition, 28.6% of CHMs occurred after 40 years of age as against 9.1% of PHMs. Findings from other works support these observations. In Egypt, 18% of HMs occurred in those above 40 years and 77.8% of these were CHM, while Banet et al. in the United States reported that all patients aged >46 years in their series had CHMs.
The ratio of CHM to PHM changes significantly with age. CHM is higher at the extremes of maternal age., Our findings in this study agree with this. Hence, comparing the percentages of CHM within the early (<25 years), mid (25–39 years), and late (≥40 years) reproductive age groups, we found that CHM is more frequent at both extremes of maternal age (42.9% and 66.7%) than the mid-reproductive age group (31.6%).
The routine use of ultrasound in pregnancy has led to the clinical diagnosis and evacuation of moles much earlier in gestation, often in the first trimester.
The mean gestational ages for all cases of HM, CHM, and PHM are 13.2 ± 3.4 weeks, 12.3 ± 3.2 weeks, and 14.7 ± 3.3 weeks, respectively. However, the mean gestational age for HM was 14.7 weeks in a previous study in Nnewi. Furthermore, two previous studies in Nigeria reported a mean gestational age of 18.6 weeks for HM and 15.2 weeks and 19.8 weeks for CHM and PHM, respectively., The findings in this study are relatively lower most probably because it is a more recent study with earlier antenatal booking and increased availability and the use of ultrasound services by pregnant women early in pregnancy.
In this study, tubal HM accounted for 5.8% of all tubal gestations, while none of the 21 CHM cases occurred in ectopic sites. This seems to agree with reports from Zaria where HM accounted for 5.1% of all tubal gestations and only 1 of 34 CHMs occurred as tubal pregnancy., In South Africa, tubal HM made up 0.23% of all tubal gestations.
Pelvic ultrasonic examination in molar gestations discloses a diagnostic snowstorm pattern, but high-resolution sonography typically reveals a complex intrauterine mass with many small cysts. This pattern, especially when associated with a markedly elevated beta-human chorionic gonadotropin (β-hCG) level, is clinically diagnostic of a molar pregnancy.
β-hCG is the most specific marker for all forms of GTD, and measurement of serum hCG levels is an integral part of the management of this disease. β-hCG is produced mainly by syncytiotrophoblasts, and it is almost invariably detectable in the serum if trophoblastic tissue is present and when sensitive assays are used.
In molar gestations, β-hCG levels at diagnosis are variable, but most show a markedly elevated β-hCG titer. Serum β-hCG levels >100,000 mIU/mL should prompt the physician to consider the diagnosis of a molar pregnancy. β-hCG titers are generally higher in complete than PHMs. For PHM, serum β-hCG levels often are in the low or normal range for gestational age. Only a few patients with PHMs show markedly elevated β-hCG titers such as those seen with CHMs. Unfortunately, due to the retrospective nature of this study, we do not have adequate data on the β-hCG levels of our cases to enable such comparisons/correlation in this study. Besides, up to 80% of the cases are first diagnosed by a histological study of spontaneously passed or curetted tissue.
The classic histologic features of CHM are villous hydrops (extensive cavitation), trophoblastic proliferation (circumferential distribution and cytological atypia), and marked cytologic atypia of intermediate trophoblast. These criteria are typically used for well-developed HM; the routine use of ultrasound in pregnancy has led to the clinical diagnosis and evacuation of moles much earlier in gestation, often in the first trimester. As a result, the classical features of CHM and PHMs that in the past were based on examination of specimens obtained in the second trimester are not as apparent, making the histopathologic diagnosis more difficult.,
In this study, the significant histologic features that were found useful in distinguishing CHM from PHM were two populations of villi, central cistern (prominent central space that is entirely acellular), circumferential trophoblastic hyperplasia, and atypia of intermediate trophoblast. Central cistern, circumferential trophoblastic hyperplasia, and atypia of intermediate trophoblast are associated with CHM. Whereas, two populations of villi are associated with PHM. However, Ishikawa et al. found that the shape of villi, predominance of villi with hydropic change, and proliferation of all three types of trophoblasts were useful to differentiate CHM from PHM. Hence, the prevalence of round and enlarged villi with proliferation of the three types of trophoblasts was a good marker for CHM; if such villi are predominantly observed with hydropic change, the case is almost certainly diagnosed as CHM. Another study reported that the significant histologic features that can distinguish CHM from PHM were central cistern, circumferential trophoblastic proliferation, and trophoblastic atypia seen in CHM versus two populations of villi, presence of fetal vessels, and scalloped border seen in PHM. The insignificant pathological morphologies that distinguished CHM from PHM in this study were scalloped outline and trophoblastic inclusions seen in PHM which were similar to previous studies.,, However, histological criteria for diagnosis of type of HM are subjective and difficult to imitate. Furthermore, histological morphology overlaps between CHM and PHM, as all features can be detected in both subtypes of HM, especially in early gestational age.
The discordant diagnosis between H and E and p57KIP2 was 14.8% which was comparable to the 4.4%–42% from other studies.,,, Compared with p57KIP2, the sensitivity and specificity of diagnosis by H and E for CHM in this study were 66.7% and 97%, respectively. The sensitivity is comparable with, while the specificity is higher than results from Popiolek et al. (53.9% and 59.2%, respectively). Gupta et al. report that the sensitivity of a diagnosis of CHM ranged from 59% to 100% for individual pathologists and the specificity ranged from 81% to 96% for individual pathologists.
The diagnostic accuracy of morphological diagnosis of CHM in this study is 85.2% and is comparable to 55%–80% reported by Vang et al. for individual pathologists, 85% published by Chen et al., and 88% obtained by Kalsoom et al.
The PPV and NPV of morphological diagnosis of CHM are 93.3% and 82.1%, respectively. This compares favorably with PPV of 88.5% and NPV of 86.6% recorded by Kalsoom et al.
From the foregoing, it can be deduced that distinction of CHM from PHM is problematic and failure to recognize all CHM by morphology persists. Gupta et al. compared the diagnostic accuracy of molar gestations between experienced gynecologic pathologists and general pathologists and concluded that the accuracy and reproducibility of a diagnosis of CHM appeared to be independent of having experience in gynecologic pathology. In this study, the inter-rater agreement between the original diagnosis and the reviewed diagnosis was very good (K = 0.86). This is particularly instructive given the fact that the reviewed diagnosis was a consensus diagnosis. However, contrary to expectation, the consensus diagnosis was not strikingly superior to the original diagnosis.
It has been suggested that problems in classification of molar specimens can be attributed to several factors, including imperfect histologic criteria for diagnosing HMs, variability in how pathologists apply diagnostic criteria, and the known variation in morphologic features dependent on the gestational age.
Previous studies have recommended an algorithm for HM diagnosis that combines the use of p57 IHC and molecular genotyping., The molecular genotyping component definitively differentiates PHM from nonmolar specimens.
However, in resource-poor settings like Nigeria where molecular genotyping laboratories are lacking or at best at its infancy, the focus should be on identifying the entity with the greatest risk for persistent GTD, i.e., CHMs. Furthermore, the misdiagnosis rate of morphological diagnosis of CHM in this study was 14.8%, while the cost of running p57 IHC is only ₦5,000.00 (about $13.7) per case. As is evident from the foregoing, the cost–benefit ratio favors routine use of p57 IHC for CHM diagnosis as against only H and E.
Hence, where IHC facilities are available, we recommend the routine use of p57 IHC staining for all POCs suspicious for HMs either clinically or morphologically. A negative p57 result establishes a CHM diagnosis with very rare exceptions, while a positive result usually rules out CHM.
It is not strictly possible to separate PHMs from hydropic abortions (HAs) – either on the basis of histomorphology or IHC staining. Although PHM tends to have higher Ki-67 IHC expression than HA, the difference is not statistically significant and hence cannot reliably distinguish PHM from HA. The evolving definition of PHM is based on chromosomal status, namely diandric (paternally derived) triploidy. Hence, it is entirely possible, even likely, that some of the cases classified as PHM in this study may actually be HA. It may therefore be safe to assume that the most important clinical distinction in this study is between CHM on the one hand versus non-CHM (PHM or HA) on the other hand.
| Conclusions|| |
HMs in general and CHMs in particular are now being diagnosed at an earlier gestational age, more often in the first trimester. Although some histological features were significant features, which differentiate between CHM and PHM, the early gestational age resulted in significant overlap in these histopathological features between the two diagnostic categories. Hence, relying on only H and E histological features to differentiate these two types of mole is fraught with suboptimal diagnostic accuracy with the potent risk of missing a good number of CHMs, some of which may progress to choriocarcinoma. In addition, the adoption of consensus diagnosis did not significantly improve the accuracy of morphology alone in the diagnosis of CHM. This study has therefore highlighted the importance of p57 immunohistochemistry in the diagnosis of this disease.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Shih LM, Ronnett BM, Mazur M, Kumar RJ. Gestational trophoblastic tumors and related tumor-like lesions. In: Kurman RJ, Ellenson LH, Ronnett BM, editors. Blaustein's Pathology of the Female Genital Tract. Cham: Springer Nature: 2019. p. 1307-75.
Gupta M, Vang R, Yemelyanova A, Kurman RJ, Li FR, Maambo EC, et al
. Diagnostic reproducibility of hydatidiform moles: Ancillary techniques (p57 immunohistochemistry and molecular genotyping) improve morphologic diagnosis for both recently trained and experienced gynecologic pathologists. Am J Surg Pathol 2012;36:1747-60.
Igwegbe A, Eleje G. Hydatidiform mole: A review of management outcomes in a tertiary hospital in South-East Nigeria. Ann Med Heal Sci Res 2013;3:210-4.
Mayun AA. Hydatidiform mole in Gombe: A five year histopathological review. Niger J Clin Pract 2008;11:134-8.
Mayun AA, Rafindadi AH, Shehu MS. Pathomorphology of molar gestation in Zaria. Niger Med J 2010;51:1-4. [Full text]
Osamor JO, Oluwasola AO, Adewole IF. A clinico-pathological study of complete and partial hydatidiform moles in a Nigerian population. J Obstet Gynaecol 2002;22:423-5.
Ekpo M. Hydatidiform mole in Nigeria. J Obstet Gynaecol 1990;10:363-6.
Audu BM, Takai IU, Chama CM, Bukar M, Kyari O. Hydatidiform mole as seen in a university teaching hospital: A 10-year review. J Obstet Gynaecol 2009;29:322-5.
Triratanachat S, Nakaporntham P, Tantbirojn P, Shuangshoti S, Lertkhachonsuk R. Role of P57KIP2 immunohistochemical expression in histological diagnosis of hydatidiform moles. Asian Pac J Cancer Prev 2016;17:2061-6.
Banet N, DeScipio C, Murphy KM, Beierl K, Adams E, Vang R, et al
. Characteristics of hydatidiform moles: Analysis of a prospective series with p57 immunohistochemistry and molecular genotyping. Mod Pathol 2014;27:238-54.
Chechia A, Koubsâa A, Makhlouf T, Anis B, Terras K, Hamouda B, et al
. Molar pregnancy. Retrospective study of 60 cases in Tunisia. Tunis Med 2001;79:441-6.
Kaye DK. Gestational trophoblastic disease following complete hydatidiform mole in Mulago Hospital, Kampala, Uganda. Afr Health Sci 2002;2:47-51.
Jun SY, Ro JY, Kim KR. P57kip2 is useful in the classification and differential diagnosis of complete and partial hydatidiform moles. Histopathology 2003;43:17-25.
Savage P, Williams J, Wong SL, Short D, Casalboni S, Catalano K, et al
. The demographics of molar pregnancies in England and Wales from 2000-2009. J Reprod Med 2010;55:341-5.
Oun AE. Clinical analysis of molar pregnancy in Egyptian population. Al Azhar Assiut Med J 2014;12:335-56.
Savage PM, Sita-Lumsden A, Dickson S, Iyer R, Everard J, Coleman R, et al
. The relationship of maternal age to molar pregnancy incidence, risks for chemotherapy and subsequent pregnancy outcome. J Obstet Gynaecol 2013;33:406-11.
Candelier JJ. The hydatidiform mole. Cell Adh Migr 2016;10:226-35.
Jimoh AA, Ajayi AB, Saidu R. Hydatidiform mole in university of Ilorin Teaching Hospital: An 8 years review. Int J Trop Med 2012;7:57-60.
Samaila MO, Adesiyun AG, Bifam C. Ruptured tubal hydatidiform mole. J Turk German Gynecol Assoc 2009;10:172-4.
van Bogaert LJ. Clinicopathological features of gestational trophoblastic neoplasia in the Limpopo province, South Africa. Int J Gynecol Cancer 2013;23:583-5.
Sebire NJ, Fisher RA, Rees HC. Histopathological diagnosis of partial and complete hydatidiform mole in the first trimester of pregnancy. Pediatr Dev Pathol 2003;6:69-77.
Merchant SH, Amin MB, Viswanatha DS, Malhotra RK, Moehlenkamp C, Joste NE. P57kip2 immunohistochemistry in early molar pregnancies: Emphasis on its complementary role in the differential diagnosis of hydropic abortuses. Hum Pathol 2005;36:180-6.
Ishikawa N, Harada Y, Tokuyasu Y, Nagasaki M, Maruyama R. Re-evaluation of the histological criteria for complete hydatidiform mole: Comparison with the immunohistochemical diagnosis using p57KIP2 and CD34. Biomed Res 2009;30:141-7.
Paradinas FJ, Browne P, Fisher RA, Foskett M, Bagshawe KD, Newlands E. A clinical, histopathological and flow cytometric study of 149 complete moles, 146 partial moles and 107 non-molar hydropic abortions. Histopathology 1996;28:101-10.
Popiolek DA, Yee H, Mittal K, Chiriboga L, Prinz MK, Caragine TA, et al
. Multiplex short tandem repeat DNA analysis confirms the accuracy of p57KIP2 immunostaining in the diagnosis of complete hydatidiform mole. Hum Pathol 2006;37:1426-34.
Crisp H, Burton JL, Stewart R, Wells M. Refining the diagnosis of hydatidiform mole: Image ploidy analysis and p57KIP2 immunohistochemistry. Histopathology 2003;43:363-73.
Vang R, Gupta M, Wu L, 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;36:443-53.
Chen KH, Hsu SC, Chen HY, Ng KF, Chen TC. Utility of fluorescence in situ
hybridization for ploidy and p57 immunostaining in discriminating hydatidiform moles. Biochem Biophys Res Commun 2014;446:555-60.
Kalsoom R, Jaffar R, Qureshi N, Aziz F. A study of p57KIP2 expression and morphological findings of complete and partial hydatidiform moles. Biomedica 2015;31:11-4.
Madi JM, Braga A, Paganella MP, Litvin IE, Wendland EM. Accuracy of p57KIP2 compared with genotyping to diagnose complete hydatidiform mole: A systematic review and meta-analysis. BJOG 2018;125:1226-33.
Alwaqfi R, Chang MC, Colgan TJ. Does Ki-67 have a role in the diagnosis of placental molar disease? Int J Gynecol Pathol 2020;39:1-7.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3]