Gestational trophoblastic disease is a group of disorders characterized by elevated β-hCG, a constellation of clinical symptoms, and may be classified into benign and malignant conditions.1 Hydatidiform mole is one of the benign conditions that may be further classified as either partial or complete mole according to microscopic findings and genetic origin.1,2 Hydatidiform mole has the ability to progress into a neoplastic or malignant condition and therefore requires an understanding of the epidemiology, an appropriate diagnosis and management, and a careful clinical follow-up.
The rate of hydatidiform mole is known to vary across different populations and groups.1,3–5,6–16 Factors which have been associated with an increased risk for hydatidiform mole include maternal age1,3–6,17–22 previous hydatidiform mole,1,3,17,18 and other modifiable and non-modifiable factors such as parity, ethnicity, socioeconomic status, and environmental exposures.1,3,17,18 However, studies characterizing rates and associated factors are dated and limited by varying definitions, and unclear or grouped diagnoses that do not distinguish between partial and complete moles6,23; there has also been poor methodological design in evaluating risk associations.1,3,17,18
This population-based study evaluated temporal trends in the rate of gestational trophoblastic disease in Nova Scotia, in relation to maternal age, using three established and reliable data sources: The Nova Scotia Gestational Trophoblastic Disease Registry, the Nova Scotia Atlee Perinatal Database, and Statistics Canada.
MATERIALS AND METHODS
Data pertaining to hydatidiform mole used in this population-based study was extracted from the Nova Scotia Gestational Trophoblastic Disease Registry, a database established in 1981 under the service of the Division of Gynecologic Oncology in the Department of Obstetrics and Gynecology at Dalhousie University, based at the Queen Elizabeth II Health Sciences Centre, Halifax, Nova Scotia. The Nova Scotia Gestational Trophoblastic Disease Registry records every case of pathologically confirmed gestational trophoblastic disease in Nova Scotia and provides patients and physicians with educational information online.24 The chosen time period, from 1990 to 2005, was selected as the most complete available data from all three analyzed databases. Anonymous data for all recorded diagnoses of hydatidiform mole from 1990 to 2005, including diagnoses of complete mole or partial mole and maternal age was extracted from the registry. The recorded diagnoses in the Nova Scotia Gestational Trophoblastic Disease Registry were all confirmed by a gynecologic pathologist. All women that were eligible for the study were included in the analysis.
The Nova Scotia Atlee Perinatal Database, overseen by the Reproductive Care Program of Nova Scotia, was established as a provincial database in 1988. It collects and maintains clinical maternal and neonatal information, such as demographic variables, procedures, maternal and newborn diagnoses and morbidity/mortality information. These data consist of all births in Nova Scotia, and to Nova Scotia residents, for pregnancies 20 or more weeks of gestational age, and for newborns 500 g or more.25 The number of total births and live births during this time period in Nova Scotia were extracted from the Nova Scotia Atlee Perinatal Database.
The data on total pregnancies in Nova Scotia was derived from the consensus data available from Statistics Canada.26 Statistics Canada incorporates data on all live births, stillborn infants and induced abortions obtained from data sources including the Canadian Birth Database (liveborn infants), Stillbirth Database (miscarriages, illegal abortions, and unspecified abortions reported in the Hospital Morbidity Databases), and the Therapeutic Abortion Survey (induced abortions performed in Canada, as well as abortions performed on Canadian citizens in prespecified states within the United States).
The data on gestational trophoblastic disease obtained from these sources was expressed as rates per pregnancy (including live births, still births, induced abortion, miscarriages and ectopic pregnancies), per total birth (including live births and stillbirths 20 or more weeks of gestational age), and per live birth, for a 16-year period in Nova Scotia. Based on the pathologic diagnosis of surgical specimens, all hydatidiform moles classified as either partial or complete moles, were considered. The effect of maternal age and year of diagnosis of disease was evaluated. Maternal age was categorized as less than 20 years, 20–29 years, 30–34 years, and greater than 35 years. The requirement for chemotherapy (including methotrexate and actinomycin) was also noted. Univariable analyses were performed using a χ2 test and Student t test, where appropriate. The Kolmogorov-Smirnov test was used to test the normality of the data on maternal age.27 Analyses were performed using SPSS 14.0 and EpiIinfo 3.2.3 (Centers for Disease Control and Prevention, Atlanta, GA). Approval for access to data from the Nova Scotia Atlee Perinatal Database was obtained from the Reproductive Care Program of Nova Scotia, and from the Research Ethics Board at the Queen Elizabeth II Health Science Centre. Approval for Nova Scotia Gestational Trophoblastic Disease Registry access was also obtained from the Research Ethics Board at the Queen Elizabeth II Health Science Centre.
The Nova Scotia Atlee Perinatal Database identified 160,843 total births and 159,972 live births during the study period, while Statistics Canada identified 163,267 total pregnancies between 1990 and 2003. The Nova Scotia Gestational Trophoblastic Disease Registry identified 428 women with gestational trophoblastic disease over the 16-year study period: 63% were partial moles, 35% were complete moles, while gestational trophoblastic neoplasm comprised 2%. Table 1 summarizes the rates for hydatidiform moles, partial moles, and complete moles. Hydatidiform moles showed rates of 220 in 100,000 (1 in 454) pregnancies, 264 in 100,000 (1 in 379) total births, and 266 in 100,000 (1 in 376) live births. Rates of partial hydatidiform mole were twofold higher than complete mole (P<.001). Figure 1 illustrates the temporal change in these rates and demonstrates no significant difference with time for partial moles (χ2; for trend, P=.11) and complete moles (χ2; for trend, P=.22); this lack of trend was observed in all of the denominators examined (ie, total pregnancy, live births, and total births). The ages did not deviate from a normal distribution for partial (P=.434) or complete (P=.284) moles. Women diagnosed with a partial mole were significantly older than those diagnosed with complete mole over all years (1990–2005) (mean age of 29 years compared with mean age of 26 years, P=.005), and there was no change in mean maternal age with time (P=.39).
Figure 2 summarizes rates of hydatidiform mole, partial mole, and complete mole by maternal age categories. The rate of partial mole was significantly higher than that of complete mole for all age ranges (P<.001), except for those women younger than 20 years old (P=.07). The ratio of partial mole to complete mole increases with increasing age: less than 20 years, 0.59; 20–29 years, 1.67; 30–34 years, 2.58; and greater than 35 years, 4.5. The χ2; for linear trend in partial moles shows a significant increase with increasing age (P<.001). Conversely the h2; for linear trend in complete moles shows a reverse trend with a significant decrease with increasing age (P<.001).
In the evaluation of hydatidiform mole per pregnancy (Fig. 2), rates were found to be significantly higher for all maternal age categories compared with those 20–29 years; younger (less than 20 years old, P=.02) and older age groups (30–34 years old, P=.04, and 35 or more years old, P=.02). There was no difference seen between the less-than-20-year, 30–34-year, and greater-than-35-year (P=.5, .52, .91) groups (Fig. 2). When rates per live births were considered (Fig. 3), hydatidiform mole was significantly more common in women younger than 20 years (P<.001) and 35 or more years of age (P=.03), but not in women 30–34 years old (P=.17), compared with women 20–29 years old. Although the rate of hydatidiform hole was more common in women younger than 20 years compared with women 30–34 years old (P=.005), no differences were noted in rates between any other groups.
In addition to surgical management of gestational trophoblastic disease, methotrexate was required in 2 (0.7%) women with a partial mole and in 29 (19.6%) women with a complete mole (P<.001); nine women (6.1%) with a complete mole also required treatment with actinomycin D.
This provincial, population-based study evaluated rates of hydatidiform mole by maternal age and year of diagnosis in Nova Scotia and estimated these rates to be higher than most previously reported rates. Our rates showed no change with time from 1990 to 2005, but did show a maternal age-related association, with higher rates demonstrated at extremes of age.
Worldwide, rates of hydatidiform mole have been variably reported per pregnancy, per total birth, and per live births. Rates in Europe have been estimated to be 83–150 per 100,000 deliveries, although somewhat higher rates have been reported in Ireland (220 per 100,000 deliveries).9 Rates as low as 66 per 100,000 pregnancies have been reported in Italy4 and in New Zealand.8 Rates in Asia are estimated to range from 81 per 100,000 to 1,754 per 100,000 pregnancies.1,3,10,11,18
In the United States, hydatidiform mole is estimated at a rate of 108–121 per100,000 pregnancies,1,3,5 although higher rates are estimated in Hawaii (460 per 100,000 live births)7 and Alaska (389 per 100,000 deliveries).28 Studies evaluating rates of hydatidiform mole in Canada are sparse and dated, estimating rates of 71 per 100,000 live births to 83 per 100,000 pregnancies.12–14
Differences in rates of hydatidiform mole may be explained by population genetic differences, environmental influences, blood type, family history, socioeconomic status,16,29 smoking,20,30 diet/nutrition, and consanguinity.6 Geographical differences (increased along the seacoast compared with inland) have also been observed.10 Our study showed a significantly higher rate of moles than previously found in Canada and the majority of the United States (220 per 100,000 pregnancies, 264 per 100,000 total births, and 266 per 100,000 live births in Nova Scotia). As mentioned previously, other studies that have reported similar rates have been based out of populations with a higher incidence of Asian, Aboriginal, or Japanese individuals.6,15,16 The 2001 Nova Scotia population census (January 2003) shows that only 3.8% of the population was a visible minority, with only 9.5% of this group Chinese and 8.4% South Asian. Only 1.9% of the total population was of Aboriginal origin. Nova Scotia, which lies on the eastern coast of Canada, has a largely Celtic heritage and a homogeneous, predominantly Caucasian population of approximately one million people.26 Rates in our study were consistent with those in Ireland.
Information on rates of hydatidiform mole may have previously been limited by considering all categories of moles as a uniform group.23 Our study was able to distinguish between categories of hydatidiform mole and demonstrated a 2:1 relationship between rates of partial mole and complete mole, consistent with one study,31 while earlier studies suggested the reverse trend.17,20,21,30 While several studies have described increasing,6,17 decreasing,32,33 and stable19 rates of hydatidiform mole over time, our study spanned a 16-year period, used reliable data, and demonstrated no temporal change in the rate of hydatidiform mole. The static rate of moles over the study period may represent a truly stable rate of disease. The other possible explanation is that the rate of molar pregnancies is in fact decreasing, but this is offset by an improving capacity to detect molar pregnancies. Given the decreasing pregnancy and birth rates in Nova Scotia, one may expect a decreasing incidence of gestational trophoblastic disease. Further follow-up of this database will be necessary to detect a true change in rate. Given the increasing average age of pregnancy that has been seen in Nova Scotia,34 the static rate of moles is likely related to the small incidence of this disease, the small numbers in this study, and the fact that the increase in maternal age is a relatively recent change in the study period.
Maternal age has consistently been shown to increase the risk of hydatidiform mole in women younger than 20 years of age and older than 35 years of age (possibly related to defects in oocyte function and formation at the extremes of reproductive age)1,3–5,15,17–19 and may be related to complete mole alone.6,20–22 An increasing contribution of partial moles to total hydatidiform moles with increasing maternal age has been postulated15 and may be due to abnormal zona pellucida formation in older women, facilitating sperm penetration and promoting formation of a partial mole. Our study also demonstrated that maternal age is a risk factor, with a significantly higher rate of hydatidiform moles in general, seen in women younger than 20 years of age, with increasing risk of partial mole with increasing maternal age. Differences were seen in rates per live births compared with women younger than 20 years of age, but not in rates per pregnancy, and may be explained by higher rates of early pregnancy losses and induced abortions in this younger age group.
Incidences have been quoted as a per pregnancy rate (ie, live births, still births, abortion, miscarriages, and ectopic pregnancies), per delivery/total birth rate (ie, live births and still births) or per live birth rate.1,6,18 The rate per live birth and per total birth is inherently flawed due to the exemption of spontaneous abortions and therapeutic abortion.1,18 This will falsely lead to an overestimation of the rate of hydatidiform mole. The rate per total pregnancies is also flawed because this may not include all the spontaneous/therapeutic abortions or home deliveries.1,6,18 Some early hydatidiform moles may be considered spontaneous abortions and be falsely included in the denominator. Total pregnancy is still the most common denominator used, but is the most difficult denominator to determine; total pregnancy is always an estimate. The numbers have been reported with three denominators in this study to easily compare these statistics to other previously performed studies.
While other studies have shown that previous hydatidiform mole, nulliparity, abortion, genetics, paternal age, parental blood group, oral contraceptives, smoking, alcohol, diet, socioeconomic status, pelvic tuberculosis, ionizing radiation, and herbicide exposure may be associated with an increased risk for hydatidiform mole,1,3,17,18,35 our study was not able to evaluate these factors.
Limitations to estimating rates of hydatidform mole have included poor study design, and inconsistent definitions and diagnoses of disease.1,6,18 The main challenge to determining rates in other studies have included the inability to account for all induced abortions/home births,1,6,10,18 and basing case diagnoses on hospital records of clinical diagnoses, fetal death certificates or systematic registries,1,3 without consistent pathologic confirmation.1,17 Other challenges included difficulties with recruitment and follow-up of large portions of the study population.18 The unique nature of the databases employed in this study allowed the determination of rates based on pregnancies, total births, and live births, information which has been lacking in previous reports. In this study, all diagnoses of gestational trophoblastic disease extracted from the Nova Scotia Gestational Trophoblastic Disease Registry were validated using the Nova Scotia Pathology Database, an electronic provincial database recording all pathologic diagnoses. Records form the Nova Scotia Gestational Trophoblastic Disease Registry accounted for 93% of diagnoses of gestational trophoblastic disease recorded in the Pathology Database, and these discordant 7% of diagnoses were unable to be confirmed as gestational trophoblastic disease when reevaluated by a gynecologic pathologist. Data collection for the Nova Scotia Atlee Perinatal Database is performed by trained health record personnel from standardized forms and medical records and data quality assurance programs carrying out regular abstraction studies are done on an ongoing basis. The Nova Scotia Atlee Perinatal Database is an accurate and reliable database.36,37
The rates of hydatidiform mole in Nova Scotia estimated by this population-based study employing comprehensive validated information are higher than most previously reported. Maternal age was a significant factor in the risk for molar pregnancies. Geographical or ethnic heritage may also be important in this population. Evaluation of other risk factors requires further assessment to ensure appropriate identification of high-risk women to optimize accurate diagnosis and clinical management of gestational trophoblastic disease.
1. Altieri A, Franceschi S, Ferlay J, Smith J, La Vecchia C. Epidemiology and aetiology of gestational trophoblastic diseases. Lancet Oncol 2003;4:670–8.
2. Lawler SD, Fisher RA, Dent J. A prospective genetic study of complete and partial hydatidiform moles. Am J Obstet Gynecol 1991;164:1270–7.
3. Steigrad SJ. Epidemiology of gestational trophoblastic diseases. Best Pract Res Clin Obstet Gynaecol 2003;17:837–47.
4. Mazzanti P, La Vecchia C, Parazzini F, Bolis G. Frequency of hydatidiform mole in Lombardy, Northern Italy. Gynecol Oncol 1986;24:337–42.
5. Hayashi K, Bracken MB, Freeman DH Jr, Hellenbrand K. Hydatidiform mole in the United States (1970–1977): a statistical and theoretical analysis. Am J Epidemiol 1982;115:67–77.
6. Smith H, Kim SJ. Epidemiology. In: Gestational trophoblastic disease. 2nd ed. Amsterdam (Netherlands): Elsevier; 2003;p. 39–79.
7. Jacobs PA, Hunt PA, Matsuura JS, Wilson CC, Szulman AE. Complete and partial hydatidiform mole in Hawaii: cytogenetics, morphology and epidemiology. Br J Obstet Gynaecol 1982;89:258–66.
8 Duff GB. Gestational trophoblastic disease in New Zealand, 1980–1986. Aust N Z J Obstet Gynaecol 1989;29:139–42.
9. Giwa-Osagie MO, Okwerekwu G. Epidemiology of molar pregnancies in Northern Ireland. Int J Gynaecol Obstet 1999;66:175–7.
10. Song HZ, Wu PC. Incidence of hydatidiform mole in China. Int J Epidemiol 1986;15:429–30.
11. Aziz MF, Kampono N, Moegni EM, Sjamsuddin S, Barnas B, Samil RS. Epidemiology of gestational trophoblastic neoplasm at the Dr. Cipto Mangunkusumo Hospital, Jakarta, Indonesia. Adv Exp Med Biol 1984;176:165–75.
12. Yuen BH, Cannon W. Molar pregnancy in British Columbia: estimated incidence and postevacuation regression patterns of the beta subunit of human chorionic gonadotropin. Am J Obstet Gynecol 1981;139:316–9.
13. Brisson J, Fabia J. Occurrence of hydatiform mole and malignant trophoblastic diseases in Eastern Quebec, 1969–1973. Union Med Can 1976;105:1238–40.
14. Henderson DL, Fraser RC. Gestational trophoblastic disease (1965–1970) Victoria General Hospital, Halifax, Nova Scotia. N S Med Bull 1972;51:9–12.
15. Tham BW, Everard JE, Tidy JA, Drew D, Hancock BW. Gestational trophoblastic disease in the Asian population of Northern England and North Wales. BJOG 2003;110:555–9.
16. Bayatpour M, Reyes J. Gestational trophoblastic disease among adolescents. Adolescent Pediatr Gynecol 1993;6:220–2.
17. Bagshawe KD, Dent J, Webb J. Hydatidiform mole in England and Wales 1973–83. Lancet 1986;2:673–7.
18. Palmer JR. Advances in the epidemiology of gestational trophoblastic disease. J Reprod Med 1994;39:155–62.
19. La Vecchia C, Parazzini F, Decarli A, Franceschi S, Fasoli M, Favalli G, et al. Age of parents and risk of gestational trophoblastic disease. J Natl Cancer Inst 1984;73:639–42.
20. Parazzini F, La Vecchia C, Pampallona S. Parental age and risk of complete and partial hydatidiform mole. Br J Obstet Gynaecol 1986;93:582–5.
21. Graham IH, Fajardo AM, Richards RL. Epidemiological study of complete and partial hydatidiform mole in Abu Dhabi: influence age and ethnic group. J Clin Pathol 1990;43:661–4.
22. Sebire NJ, Foskett M, Fisher RA, Rees H, Seckl M, Newlands E. Risk of partial and complete hydatidiform molar pregnancy in relation to maternal age. BJOG 2002;109:99–102.
23. Szulman AE, Surti U. The syndromes of hydatidiform mole. I. Cytogenetic and morphologic correlations. Am J Obstet Gynecol 1978;131:665–71.
24. Nova Scotia Gestational Trophoblastic Disease Registry. Available at: http://gynoncology.medicine.dal.ca/Pages/GTD.html
. Retrieved January 1–20, 2007.
25. Reproductive Care Program of Nova Scotia. Available at: http://rcp.nshealth.ca
. Retrieved December 18, 2006.
26. Stats Canada. Available at: http://cansim2.statcan.ca
. Retrieved January 14, 2007.
27. Riffenburgh RH. Statistics in medicine. 2nd edition. Philadelphia (PA): Elsevier Academic Press; 2006.
28. Martin PM. High frequency of hydatidiform mole in native Alaskans. Int J Gynaecol Obstet 1978;15:395–6.
29. Talati NJ. The pattern of benign gestational trophoblastic disease in Karachi. J Pak Med Assoc 1998;48:296–300.
30. La Vecchia C, Franceschi S, Parazzini F, Fasoli M, Decarli A, Gallus G, et al. Risk factors for gestational trophoblastic disease in Italy. Am J Epidemiol 1985;121:457–64.
31. Jeffers MD, O’Dwyer P, Curran B, Leader M, Gillan JE. Partial hydatidiform mole: a common but underdiagnosed condition. A 3-year retrospective clinicopathological and DNA flow cytometric analysis. Int J Gynecol Pathol 1993;12:315–23.
32. Matsui H, Iitsuka Y, Yamazawa K, Tanaka N, Seki K, Sekiya S. Changes in the incidence of molar pregnancies: a population-based study in China prefecture and Japan between 1974 and 2000. Hum Reprod 2003;18:172–5.
33. Kim SJ, Bae SN, Kim JH, Kim CJ, Han KT, Chung JK, et al. Epidemiology and time trends of gestational trophoblastic disease in Korea. Int J Gynaecol Obstet 1998;60:S33–8.
34. Joseph KS, Young DC, Dodds L, O’Connell CM, Allen VM, Chandra S, Allen AC. Changes in maternal characteristics and obstetric practice and recent increases in primary cesarean delivery. Obstet Gynecol 2003;102:791–800.
35. Loukovaara M, Pukkala E, Lehtovirta P, Leminen A. Epidemiology of hydatidiform mole in Finland, 1975 to 2001. Eur J Gynaecol Oncol 2005;26:207–8.
36. Fair M, Cyr M, Allen AC, Wen SW, Guyon G, MacDonald RC. An assessment of the validity of a computer system probabilistic record linkage of birth and infant death records in Canada. Chronic Dis Can 2000;21:8–13.
37. Fair M, Cyr M, Allen AC, Wen SW, Guyon G, MacDonald RC. Validation study for a record linkage of births and infant deaths in Canada. Catalogue No. 84F0013XIE. Ottawa: Statistics Canada; 1999.