Changes in the incidence of molar pregnancies. A population-based study in Chiba Prefecture and Japan between 1974 and 2000

Hideo Matsui1, Yoshinori Iitsuka, Koji Yamazawa, Naotake Tanaka, Katsuyoshi Seki and Souei Sekiya

Department of Obstetrics and Gynecology, Chiba University School of Medicine, 1–8-1 Inohana, Chuo-ku, Chiba 260–8670, Japan 1 To whom correspondence should be addressed. e-mail: hmatsui{at}med.m.chiba-u.ac.jp


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: The aim of this population-based retrospective study was to evaluate the trend in the incidence of molar pregnancy. METHODS: Questionnaires about the occurrence of hydatidiform mole were mailed to all clinics and hospitals in Chiba Prefecture between 1974 and 2000. The diagnosis of hydatidiform mole was based on the post-operative morphological and/or pathological findings. Data of live births were collected from the annual reports of maternal and child health statistics. A total of 3778 women with hydatidiform mole and 1 661 579 women with live births were analysed. RESULTS: The incidence of hydatidiform mole per 1000 live births was constant from 1974 to 1990, while it decreased significantly after 1991 ({chi}2-test for trend, P < 0.0001). The incidence of complete mole also decreased after 1991 ({chi}2-test for trend, P = 0.0001), though the incidence of partial mole remained unchanged. CONCLUSIONS: The incidence of hydatidiform mole in Chiba Prefecture has become as low as that in Europe or the USA, and it is attributed to the decreasing incidence of complete mole. A number of environmental factors that prevent the ovulation of ‘empty’ oocytes might have contributed to the decreasing incidence of complete mole.

Key words: hydatidiform mole/incidence/population-based study


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Molar pregnancy is classified into two entities, complete and partial moles, on the basis of morphological and cytogenetic examinations (Vassilakos et al., 1977Go; Szulman and Surti, 1978Go). Genetically, complete mole (CM) has been shown to be the consequence of an ‘empty’ oocyte (i.e. one that has a degenerative nucleus) being fertilized by a haploid sperm or two sperm, carrying a karyotype of 46,XX or 46,XY of purely paternal origin (androgenesis) (Kajii and Ohama, 1977Go; Ohama et al., 1981Go). On the other hand, the majority of partial moles (PMs) are formed by fertilization of an apparently normal oocyte by two sperm, and the karyotype is usually triploid (69,XXY or 69,XXX) (Lawler et al., 1982Go; Ohama et al., 1986Go).

The incidence of molar pregnancy demonstrated marked geographic and ethnic differences, ranging from the highest incidence of 1 in 120–400 pregnancies in Asian countries such as Taiwan, Philippines and Japan, to the lowest incidence of 1 in 1000 to 2000 in Europe and the USA (Bracken et al., 1984Go). These 10–20-fold variations in the incidence of molar pregnancy might be overestimated by reporting biases, such as population-based and hospital-based data. Recently, the incidence of molar pregnancy in Japan was reported to be only two or three-fold higher than that in the USA on the population-based study (Hayashi et al., 1982Go; Takamizawa et al., 1987Go; Ishizuka and Tomoda, 1990Go).

The population-based registration system for molar pregnancy was established in 1974 in Chiba Prefecture together with 12 other prefectures in Japan (Takamizawa et al., 1987Go; Ishizuka and Tomoda, 1990Go). This registration system has been introduced in more regions, and now this population-based registration system is adopted in 22 of the 46 prefectures in Japan. We have analysed trends in the incidence for hydatidiform mole in Chiba Prefecture and 22 prefectures of Japan over a period of 27 years.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Chiba Prefecture, located next to Tokyo, had a population of ~ 4 140 000 in 1974. The population has constantly increased to 5 830 000 in 1998 and the vast majority is Japanese. On the contrary, the number of births has constantly decreased from 82 078 in 1974 to 55 318 in 2000. There were 359 clinics and hospitals with obstetric and gynaecological services in 1974 and in 2002 there were 293 clinics and hospitals. Questionnaires have been mailed every year to all clinics and hospitals in Chiba Prefecture to assess whether or not new patients with hydatidiform mole have been diagnosed. Approximately 90–95% of questionnaires were returned every year and a total of 3778 patients with hydatidiform mole have been registered during the past 27 years (1974–2000).

Since 1985, we have established a new system to register CM and PM, separately. The diagnostic criteria of CM and PM were based on the morphological findings and/or pathological findings at evacuation. Complete mole was diagnosed as the presence of diffuse hydropic vesicles and the absence of identifiable fetal parts. In contrast, PM was diagnosed as the presence of both normal and hydropic villi with or without fetal parts. The morphological diagnostic criteria for hydatidiform mole (hydropic vesicles >2 mm in diameter) have not changed for the 27 year study period.

A similar population-based registration system of hydatidiform mole has been carried out in 22 prefectures in Japan. The outcomes of this registration system in Japan (from 1976 to 1997) were compared with those in Chiba Prefecture.

Statistical analysis
Statistical analyses were performed by Welch’s t-test, {chi}2-test for heterogeneity and the {chi}2-test for trend (Fisher and Van Belle, 1993Go).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Annual number and incidence of molar pregnancy and postmolar trophoblastic tumour
Table I shows the annual number and incidence of molar pregnancies per 1000 live births in Chiba Prefecture during 1974–2000. The number of molar pregnancies in 1974 was 205 cases and this has gradually declined to just 91 cases during 2000. The number of live births has also decreased from 82 078 in 1974 to 55 318 in 2000. The incidence of molar pregnancy per 1000 live births remained constant from 1974–1990 and was two- or three-fold higher than that in Europe and the USA. In contrast, the incidence decreased in 1991 compared with the 1974–1990 period ({chi}2-test for heterogeneity, P = 0.001) and is now as low as that of Europe and the USA. The overall rate of postmolar gestational trophoblastic tumour was 9.4% (342/3778) and appears to have been constant for the past 27 years.


View this table:
[in this window]
[in a new window]
 
Table I. Incidence of molar pregnancies per 1000 live births and persistent trophoblastic tumour in Chiba Prefecture between 1974 and 2000
 
Figure 1 shows the trends in the median incidence of molar pregnancy per 1 000 live births in Chiba Prefecture and 22 other prefectures in Japan. The incidence per 1000 live births has declined from 2.50 in 1974 to 1.65 in 2000 in Chiba Prefecture ({chi}2-test for trend, P < 0.0001) and from 2.79 in 1976 to 1.61 in 1997 in Japan respectively.



View larger version (10K):
[in this window]
[in a new window]
 
Figure 1. Incidence of hydatidiform mole per 1000 live births (•) in Chiba Prefecture from 1974 to 2000, and incidence in 22 Prefectures of Japan from 1977 to 1997 (X).

 
The annual number and incidence of complete and partial mole
Since 1985, we have registered CM and PM separately according to macroscopic and/or microscopic findings. A total of 992 patients with CM and 717 patients with PM have been registered during these 16 years. Table II shows the annual number and incidence of CM and PM per 1 000 live births in Chiba Prefecture from 1985 to 2000. The mean (±SD) incidence of CM in the period 1985–1990 was 1.71 (± 0.17) and it remained constant while the incidence of CM significantly decreased in 1991 compared with that in the previous period ({chi}2-test for heterogeneity, P = 0.0008). Moreover, the incidence of CM has been further decreasing from 1.09 in 1991 to 0.49 in 2000 ({chi}2-test for trend, P = 0.0001). In contrast, the incidence of PM has stayed constant throughout these 16 years.


View this table:
[in this window]
[in a new window]
 
Table II. Annual number and relative incidences of complete and partial mole in Chiba Prefecture between 1985 and 2000
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The two or three-fold higher incidence of molar pregnancy in the countries of South-East Asia than that of the USA or Europe has been reported on the population-based study (Hayashi et al., 1982Go; Takamizawa et al., 1987Go; Ishizuka and Tomoda, 1990Go). Although several environmental factors such as protein deficiency, malnutrition and low socio-economic status have been proposed as contributing to this geographic and ethnic distribution, definite environmental factors are not demonstrated.

In this study, the incidence of molar pregnancy in Chiba Prefecture has been constant from 1974 to 1990, as observed in our previous study (Takamizawa et al., 1987Go), while a significant decline has been demonstrated since 1991. A similar decreasing tendency has also been observed in Japan as a whole. Changes in the incidence of molar pregnancy seem to result from the decline of complete mole, and now the incidence of molar pregnancy in Chiba Prefecture has become as low as that in Europe or the USA.

It has also been reported (Martin and Kim, 1998Go) that the median incidence of molar pregnancy per 1000 live births in Korea had decreased over the past four decades from 4.4 in the 1960s to 1.6 in the 1990s. They considered that the decreasing incidence was due to changing social conditions, the decreased percentage of older women and those of higher parity in the obstetric population in Korea. Maternal ages >40 years old were generally accepted as being a high risk factor for developing complete mole. On the contrary, the percentage of deliveries at an older maternal age (>40 years) in Japan has slightly increased (0.45% in 1980, 1.0% in 1990, and 1.1% in 1998) (Ministry of Health and Welfare, 2000).

Other possible causes for the recent decrease in the incidence of molar pregnancy may be either missed or underestimated complete mole, which may be diagnosed as partial mole or non-molar spontaneous abortion based on macroscopic examinations (Soto-Wright et al., 1995Go; Mosher et al., 1998Go; Lindholm and Flam, 1999Go). Recent advances of high sensitivity pelvic ultrasonography may have led to early termination of abnormal pregnancies, before we can detect classical symptoms or identify hydatidiform vesicles macroscopically. At our hospital, the mean gestational ages at evacuation of patients with CM and PM were 11.0 ± 3.0 weeks (n = 285) and 11.0 ± 3.1 weeks (n = 55) in the former period (1985–1990), and 9.9 ± 2.2 weeks (n = 176) and 9.5 ± 2.0 weeks (n = 133) in the latter period (1991–2000) respectively. The mean gestational ages at evacuation of patients with CM and PM were significantly shortened in the latter period (P < 0.05; Welch’s t-test).

It has been previously reported (Jacobs et al., 1982Go) that Japanese people in Hawaii who had acclimatized to an American lifestyle for a long period appeared to lose the increased risk of molar pregnancy. Although the reason for this change has not been entirely clear, an improvement of several environmental factors such as vitamin A deficiency, decreasing levels of consumption of dietary carotene and animal fat, and low socio-economic status have been reported to contribute to the decrease of molar pregnancy.

In this study, the ethnic background of the Japanese is relatively homogeneous and stable, and the diagnostic criteria for CM and PM have remained unchanged for these two decades. Moreover, the percentage of deliveries at an older maternal age, with a high risk factor for developing complete mole, is increasing in Japan. However, the incidence of complete mole per 1 000 live births has demonstrated a significant decrease in 1991 compared with that in the previous period (P = 0.0008), and it has been further decreasing from 1.09 in 1991 to 0.49 in 2000 (P = 0.0001). In contrast, the incidence of PM has stayed constant during these16 years. Since the genesis of complete hydatidiform mole results from fertilization of an empty oocyte by a haploid sperm or by two sperm (Kajii and Ohama, 1977Go; Ohama et al., 1981Go), it could be speculated that Japanese women of reproductive age may have lost, over the last decade, the increased risk of ovulating empty oocytes.

In conclusion, the incidence of hydatidiform mole and complete mole has significantly decreased in Chiba Prefecture during the last decade. Some environmental or socio-economic factors have been proposed to contribute to the decrease in CM rates. Although we cannot clarify these aetiologic factors at the moment, a prolonged period of acclimatization to a Western lifestyle could be a factor responsible for a change in female reproductive physiology.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Bracken, M.B., Brinton, L.A. and Hayashi, K. (1984) Epidemiology of hydatidiform mole and choriocarcinoma. Epidemiol. Rev., 6, 52–75[ISI][Medline]

Fisher, L.D. and Van Belle, G. (1993) Biostatistics: a methodology for the health sciences. John Wiley & Sons, Inc. New York, pp. 253–255.

Hayashi, K., Bracken, M.B., Freeman, Jr., D.H. and Hellenbrand, K. (1982) Hydatidiform mole in the United States (1970–1977): A statistical and theoretical analysis. Am. J. Epidemiol., 115, 67–77.[Abstract]

Ishizuka, N. and Tomoda, Y. (1990) Gestational trophoblastic disease: hydatidiform mole, invasive mole and choriocarcinoma. The University of Nagoya Press, Nagoya, pp. 1–8.

Jacobs, P.A., Hunt, P.A., Matsuura, J. and Wilson, C. (1982) Complete and partial hydatidiform mole in Hawaii: cytogenetics, morphology and epidemiology. Br. J. Obstet. Gynecol., 89, 258–266.

Kajii, T. and Ohama, K. (1977) Androgenetic origin of hydatidiform mole. Nature, 268, 633–634[ISI][Medline]

Lawler, S.D., Fisher, R.A., Pickthall, V.J., Povey, S. and Evans, M.W. (1982) Genetic studies on hydatidiform moles. I. The origin of partial moles. Cancer. Genet. Cytogenet., 5, 309–320.

Lindholm, H. and Flam, F. (1999) The diagnosis of molar pregnancy by sonography and gross morphology. Acta Obstet. Gynecol. Scand., 78, 6–9.[CrossRef][ISI][Medline]

Martin, B.H. and Kim, J.H. (1998) Changes in gestational trophoblastic tumors over four decades: a Korean experience. J. Reprod. Med., 43, 60–68.[ISI][Medline]

Ministry of Health and Welfare (2000) Maternal and Child Health Statistics of Japan. Mothers’ & Children’s’ Health Organization. Tokyo, pp. 40–50.

Mosher, R., Goldstein, D.P., Berkowitz, R., Bernstein, M. and Genest, D.R. (1998) Complete hydatidiform mole. Comparison of clinicopathologic features, current and past. J. Reprod. Med., 43, 21–27.[ISI][Medline]

Ohama, K., Kajii, T., Okamoto, E. Fukuda, Y., Imaizumi, K., Tsukahara, M., Kobayashi, K. and Hagiwara, K. (1981) Dispermic origin of XY hydatidiform moles. Nature, 292, 551–552.[CrossRef][ISI][Medline]

Ohama, K., Ueda, K., Okamoto, E., Takenaka, M. and Fujiwara, A. (1986) Cytogenetic and clinicopathologic studies of partial moles. Obstet. Gynecol., 68, 259–262.[Abstract]

Soto-Wright, V., Bernstein, M., Goldstein, D.P. and Berkowitz, R.S. (1995) The changing clinical presentation of complete molar pregnancy. Obstet. Gynecol., 86, 775–779.[Abstract/Free Full Text]

Szulman, A.E. and Surti, U. (1978) The syndromes of hydatidiform mole. Morphologic evolution of the complete and partial mole. Am. J. Obstet. Gynecol., 132, 20–27.[ISI][Medline]

Takamizawa, H., Matsui, H. and Inaba, N. (1987) Epidemiology of gestational trophoblastic disease. In: Takagi, S., Friedberg, V., Haller, U., Knapstein, V.P.G., Sevin, B.V. (eds), Gynecologic Oncology, Surgery and Urology. Central Foreign Books, Tokyo, pp. 236–240.

Vassilakos, P., Riotton, G. and Kajii, T. (1977) Hydatidiform mole: two entities. a morphologic and cytogenetic study with some clinical considerations. Am. J. Obstet. Gynecol., 127, 167–170.[ISI][Medline]

Submitted on May 16, 2002; Accepted on September 9, 2002