Pathomorphological Characteristics of Trophoblast and Serum Human Chorionic Gonadotropin Levels in Diagnosis of Partial Hydatidiform

© 2013 Gordana Bogdanoviæ et al.; licensee University of Sarajevo Faculty of Health Studies. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. UNIVERSITY OF SARAJEVO FACULTY OF HEALTH STUDIES ABSTRACT


INTRODUCTION
Hydatidiform mole is a condition characterized by trophoblast tissue altered into numerous and al-most transparent vesicles of various size partially or completely replacing normal chorionic villi.Pathologically changed villi are interconnected by thin but strong layers of connective tissue so macroscopically the placenta appears as fi sh roe-like or as clusters that resemble grapes -hydatidiform mole (1).Based on diff erences in morphology, histopathology, karyotype and clinical features, hydatidiform mole can be categorized into partial and complete moles.A complete mole (55%-75% of all molar pregnancies) is characterized by hydropically degenerated villi, absence of embryo and amniotic sac.A partial mole (25%-45% of all molar pregnancies) is characterized by a partial degeneration of villi while trophoblast proliferation is focally pronounced (2).Hydatidiform mole incidence varies from 0.5 to 8.3 per 1000 live born children and is signifi cantly different across countries.Th e incidence of molar pregnancy in Asia is seven to ten folds higher than in the countries of North America and Europe (3).Statistically, 290 cases of pathological trophoblast a year per 300 000 births and miscarriages were reported in Croatia (4).1.5 cases of the gestational trophoblastic disease per 1000 births was reported during a seven-year-period at the Department of Gynaecology and Obstetrics, Clinical Centre Serbia (5).In our country, however, the incidence of gestational trophoblastic disease cannot be precisely reported since there is neither a register of disease nor a disease specifi c program and management protocol.Since β HCG is synthesized in syncytiotrophoblast, it is a signifi cant indicator of its functional condition.Due to trophoblast proliferation, ever present in molar chorionic villi, β HCG titer is increased although absolute titer value by itself is not a reliable evidence of hydatidiform mole presence (1).Normal trophoblast diff erentiation and function during an implantation and placentation are of great importance for a successful pregnancy, while disorders in those processes contribute to development of numerous pathologic pregnancies including the gestational trophoblast disease (6).Abnormal trophoblast hyperplasia is a requirement for the diagnosis of molar pregnancy (7).In practice, it has been observed that correlation between histological appearance of mole and its clinical course has not been constant and absolute.Th e well diff erentiated mole can have a malignant course while anaplastic mole can be innocent (8).According to recent studies, most of the authors found numerous complications in molar pregnancy (9).Molar pregnancies are considered as premalignant lesions since they can be malignantly altered.Approximately 15%-25% of mole develops into an invasive mole and 3%-5% into choriocarcinoma (10).Th e gestational choriocarcinoma is preceded by a hydatidiform mole in 30% to 60% patients which is 1000 times higher than after a normal pregnancy (8).
Th e aim of this study was to investigate the importance of determining serum human chorionic gonadotropin levels as well as the importance of pathomorphologic analysis of trophoblast changes as a source of β HCG with a goal of using those methods in diagnostics.

Participants and Methods
Th is was a retrospective study which included 70 pregnant women diagnosed with hydatidiform mole or with physiological pregnancies spontaneously aborted.Based on survey results and a patient's fi le, diagnostic test data were processed while fi ndings were pathohistologically verifi ed at the Department of Pathology, UCC Tuzla.Th e pregnant women were reported to have almost similar demographic characteristics and were included in two groups.35 pregnant women with a molar pregnancy diagnosed during the fi rst trimester treated by evacuating the molar tissue by uterine suction or curettage were included in the study group.35 pregnant women with physiological pregnancy spontaneously aborted during the fi rst trimester treated by uterine suction or curettage were included in the control group.Molar pregnancy diagnosis was suggested by detailed patient history; gynaecological and ultrasound examination; serum β HCG level and pathohistological tissue verifi cation after an evacuation of the uterine cavity.Inclusion criteria: 1) singleton pregnancy, 2) gestational age until the 12 th week (the fi rst trimester), 3) reliable gestational age (exact date of the last menstrual period, early ultrasound examination), 4) molar pregnancy diagnosis, 5) physiological pregnancy terminated by miscarriage due to cervical insufficiency.Specifi c serum β HCG levels were noticed in all pregnant women (study and control group) before a pregnancy termination, which were determined by a quantitative β HCG assay (the Architect total β HCG) using ARCHITECT CI 800.Blood samples were extracted from a cubital vein according to a standard procedure.Available data were compared with reference values for a gestational age (11).Conception tissue obtained after the suction curettage of the cavum uteri was fi xed in buff ered formaldehyde solution (pH 7.2-7.4),paraffi n embedded, while 4μm tick histological sections were stained by haematoxylin and eosin method to examine the basic light-microscopic morphological characteristics.

Statistical analysis
Derived values were processed by standard statistical methods such as calculation of mean and standard deviation or median and interquartal range depending on data distribution.Th e chi-square test was used for determining diff erences in distribution of cross-section qualitative variable (independent distribution).ANOVA was used to test equality of arithmetic mean of quantitative variable and a factor.Th e results are shown in tables and graphs but also in clear written presentation with numerical analysis.Standard level of signifi cance p<0.05 was chosen as the statistical signifi cance and non-parametric statistical tests the Mann Whitney test, X ² test and Fisher test were also used for evaluation.

Serum β HCG Values
Th e average β HCG levels are shown in Table 1 and Figure 1.Th e average β HCG levels in the study group was 60191.37 mIU/L and in the control group 2021.76 mIU/L.Based on the standard level of signifi cance p< 0.05 with the risk of 5%, it can be concluded that there was a statistically signifi cant diff erence in the average β HCG levels between the groups.Th erefore, the higher average β HCG level was recorded in the study group.Signifi cantly higher β HCG levels were recorded in pregnant women with hydatidiform mole at all gestational ages.

Pathomorphological Characteristics of Trophoblast
Microscopic characteristics of trophoblastic chorionic villi in partial mole were examined following the molar pregnancy guidelines according to Genest (12).Pathomorphological characteristics were present only in the study group.Th e results of examination of types of atypia, mass and amount of trophoblast proliferation are shown in Table 2. Th e prevalence of mild atypia was observed in 57,14% of cases, analyzing the trophoblast atypia incidence.Statistically signifi cant diff erence in proportional prevalence of individual types of trophoblast proliferation mass was not observed.Th e most prevalent was the pronounced trophoblast proliferation mass in 45.71% of cases.Examining the amount of villi aff ected by trophoblast proliferation, the moderate trophoblast proliferation amount was observed in 51.43%, which was signifi cantly higher in contrast to the mild trophoblast proliferation amount 42.86%.Th e Figure 2 illustrates that among the most signifi cant histopathological villous changes in partial hydatidiform mole more than 1/2 cases had high β HCG levels.

DISCUSSION
Th e most important part of routine preoperative diagnostic screening of molar pregnancy is a quantitative HCG level measurement as well as measurement of its subunit β.Th e hyperplastic trophoblastic epithelium, either normal or atypical, produces increased β HCG levels.Th e levels exceed those recorded in an early pregnancy and are substantially above 50000 IU.Th e upper limit of β HCG values is not determined and can amount to several hundred thousand IU/L (13).Th e β HCG secretion is proportional to the amount of viable trophoblast (14).Th e literature suggests to variability in β HCG levels in molar pregnancy.Genest et al. (15) pointed to the preevacuation β HCG level over 100000 IU in 46% cases.Menczer (16) reported that 30 (41%) of 74 patients with molar pregnancy showed the preevacuation β HCG level over 100000 IU.Th e present study results suggest that the average β HCG serum level in patients with molar pregnancy was 60191.37 ± 49662,75 which was statistically significantly higher in comparison with normal pregnancy (2021.76 ± 2974.73).Th ose obtained results are in compliance with the results of other studies.In his study Ben Temime (17) examined 90 cases of molar pregnancy and noticed that β HCG serum levels ranged from 20000 IU/L to 40000 IU/L.Trissy (18) described a patient who was at 12 th week of gestational age and who had β HCG level of 59540 IU.Partial hydatidiform mole is associated with lower β HCG levels in comparison with complete hydatidiform mole.Th e Soto-Wright study (19) reported the preevacuationa β HCG level over 100000 IU/L in only 6% of patients with the partial hydatidiform mole, and Berkowitz (20) also reported in 2 (6%) patients of 30.Czernobilski et al. (21) reported the preevacuation urinary level over 300000 IU/L in one (6%) of 17 patients with the partial hydatidifrom mole.In his study Chechia (22) reported even 91.4% of cases with β HCG level over 50000IU/L in 60 examined molar pregnancies.β-HCG levels in molar pregnancy do not double as they should every two days but they are much higher than in a normal pregnancy.Th erefore, dynamic increase of titer should be monitored during pregnancy especially in the postevacuation period.Our results also suggest that β-HCG serum measurements are vital and reliable parameter for accurate diagnosis of hydatidiform mole.High β HCG levels were recorded in all statistically most significant pathomorphological vilous changes.Its levels as well as dynamic oscillations in serum values are the key parameters for disease severity scoring and further treatments.Th e basic functional unit of placenta contains chorionic villi, fi nger-like projections of chorion, and terminal villi represent end branches of villous tree (23).Th ere is epithelium (trophoblast) on the chorionic villi surface which consists of two clearly defi ned layers until the fi rst trimester -syncytiotrophoblast and cytotrophoblast.Disorder in regulation of blastocyst invasion is associated with the most pathological pregnancies (24).Partial mole appears as a mosaic of normal and pathologically changed villi, it is characterized by the existence of a mixture of various villi population consisting of morphologically normal villi and edematous ones of irregular shape that have cisterns and trophoblastic hyperplasia (7).Molar pregnancy is characterized by various degree of hyperplasia and anaplasia of chorionic epithelium (25).Trophoblast proliferation was present in all examined mole.Mild trophoblast proliferation typically matching a gestational age was noticed in a normal pregnancy.However, it was not considered to be pathological.Abnormal, nonpolar trophoblast hyperplasia was present in molar pregnancy and it was almost always local and less pronounced then in complete mole.It was usually multifocal rather than circumferential showing a lace-like patter or vacuolar appearance resembling cell cavities (7).Presence of trophoblast proliferation in partial mole was suggested by numerous authors (7,12,26).Unlike a normal pregnancy, presence of trophoblast atypia was observed in all cases with partial mole.Mild atypia was statistically most signifi cant (57.14%) in comparison with moderate (28.57%) and pronounced (14.29%).Montes (27) in his study on trophoblast atypia incidence reported the focal atypia in 5% cases out of 22 spontaneously aborted pregnancies, predominantly focal in 40% of 30 partial mole out of which 33% moderate and 7% pronounced, and predominantly diff use in 87% of 47 cases of complete mole.Abnormal trophoblast hyperplasia is a requirement for the diagnosis of molar pregnancy (7).An atypical patter of trophoblast hyperplasia with peripheral or multifocal pattern rather than the polar accentuation seen in a normal fi rst trimester placenta seems to be the important diagnostic feature for partial mole (28).Examining trophoblast proliferation mass in our study, we noticed that the pronounced one prevailed in 45.71% of cases, while moderate and mild were observed in 28.57% and 25.71% of cases respectively.Th e amount of villi aff ected by trophoblast proliferation was the most signifi cant in moderate (51.43%).Salafi a (29) reported the highest percentage of moderate mass and amount of trophoblast proliferation.Of 20 examined partial mole, Jaff ar (30) reported focal and diff use trophoblast proliferation in 75% and 15% cases respectively.Th e results on the presence of variability of villous trophoblast were also reported by the authors (26,31).Th e presence of three types of trophoblast proliferation in molar pregnancy was suggested by Ishikawa (32).Th ere are great diff erences in mass and amount of trophoblast proliferation atypia in certain pathological pregnancies.Supporting the idea, Park and Lees in 1950 quoted: "Morphologically, trophoblast with benign future is completely similar to trophoblast with malignant future" (25).Th erefore, the material obtained by curettage after the molar tissue has been evacuated is of a great importance in diagnostics.Th e tissue is the most signifi cant evidence of the extent to which the chorionic epithelium invaded a wall of the uterus and blood vessels indicating damaging eff ects as well as clinical potential of mole (25).Hydatidiform mole with pronounced trophoblastic cells hyperplasia indicates to a precancerous condition, i.e. a state associated with a signifi cantly increased risk of cancer.Choriocarcinoma occurs to ten times more frequently after the pregnancy with hydatidiform mole then in normal pregnancy (1).Risk of developing cancer is six times lesser with partial mole then it is seen with complete mole (10).Since we recognized the typical microscopic trophoblast characteristics of molar pregnancy according to (12), it can be concluded that the pathomorphologic analysis of evacuated tissue is a reliable indicator as well as a gold standard for partial mole diagnostics.

CONCLUSION
β-HCG level measurements and pathomorphological analysis of trophoblast changed villi are signifi cant in diagnosis of early disease stages enabling making the right treatment decisions as well as reducing morbidity and mortality.

FIGURE
FIGURE 1.Average β HCG levels in the groups according to a gestational age

FIGURE 2 .
FIGURE 2. Relationship between the most signifi cant pathomorphological changes and β HCG levels

1 .
Average β HCG levels in the groups according to a gestational age

TABLE 1 .
Difference in average β HCG levels between groups

TABLE 2 .
Number and Characteristics of Pregnant Women according to Pathomorphologic Trophoblast Characteristics