Childbirth as Protective of the Health of Women in Contrast to Induced Abortion. (I.) Breast, Ovarian, and Endometrial Cancer


Endometrium (Uterus) Cancer

The American Cancer Society estimates that there will be 34,900 cases of cancer of the uterine corpus of the uterus, usually of the endometrium, among U.S. women in 1997 and that 6000 women will die from it in 1997. However, the relationship between endometrial cancer and reproductive history, parity, or induced abortion has been little studied, but the findings that childbirth is protective and incomplete pregnancies are not protective have been consistent. The following represent virtually the only published studies in the last 20 years, which can be found in the medical literature.

Cancer Facts & Figures-1997, American Cancer Society: Atlanta, GA (1997).

Estrogen is the major risk for one type of endometrial cancer. Estrogen-related exposures include never having children, and a history of failure to ovulate have been shown to increase risk. Pregnancy and the use of oral contraceptives appear to provide protection against endometrial cancer.


Comment: In light of the available data, it appears that it is childbirth and not pregnancy, which protects against endometrial cancer.

Is the Risk of Cancer of the Corpus Uteri Reduced by a Recent Pregnancy? A Prospective Study of 765,756 Norwegian Women, G Albrektsen et al., Int'l J. Cancer 61: 485-490, 1995.

A large study of 765,756 Norwegian women representing a total of 9,307,118 persons in the age interval of 30-56 years, was undertaken using various registries. Compared to women with one full-term pregnancy (1.0), nulliparous women had a 1.94 (1.46-2.59) incidence rate ratio (IRR), women with two full-term pregnancies had an 0.85 (0.64-1.09) IRR, women with 3 full-term pregnancies had a 0.61 (0.46-0.82) IRR, women with 4 or more full-term pregnancies had a 0.48 (0.34-0.69) IRR for endometrial carcinoma. The reduction in risk was found to be more pronounced with the first pregnancy than that observed for any subsequent pregnancy. The risk of endometrial carcinoma increased with increasing time since last birth. The reduction in risk among parous women compared to nulliparous women diminished with increasing time since last birth. They concluded that "our results support the hypothesis that the reduction in risk of endometrial carcinoma associated with a pregnancy is related to a mechanical shed of malignant or pre-malignant cells at each delivery."


Comment: In Norway, there is a vast array of registries available to researchers, which enabled the study to be done. The conclusion that delivery of a child helps shed malignant or pre-malignant cells and thus reduces the likelihood of endometrial cancer was an important one, which had been hypothesized earlier by other researchers.

A Case-Control Study of Endometrial Cancer in Relation to Reproductive, Somatometric, and Life-Style Variables, A Kalandidi et al., Oncology 53: 354-359, 1996.

A hospital-based case-control study of cancer of the endometrium was conduced in Athens, Greece, from 1992-94 by researchers at the University of Athens Medical School and the Harvard School of Public Health. It was found that the risk of endometrial cancer consistently decreased with the number of live births but did not decrease with one miscarriage or one induced abortion. Compared with never-pregnant women as the base reference (1.0), women who had been pregnant with no live birth had a 1.09 OR (0.28-4.19).

Reproductive, menstrual, and medical risk factors for endometrial cancer: Results from a case-control study, LA Brinton et al., Am J. Obstet Gynecol 167:1317-1325, 1992.

During 1987-90, a study was undertaken of 405 cases of newly diagnosed cancer of the uterine corpus in women between the ages of 20-74 years, which were obtained from 7 hospitals throughout the United States. Population controls (297) were matched for age, race and location of residence, and were obtained by random digit-dialing techniques. The mean age of the cases at interview was 59.2 years, compared to 58.0 years for controls. Compared to women with no term births (1.0), women with term births had a consistently lower risk of endometrial cancer (Table lIl).


Table III
Risk of Endometrial Cancer with
Number of Term Births
Number of Term BirthsRelative Risk
none1.00
10.55 (0.3-1.1)
20.32 (0.2-0.6)
30.42 (0.2-0.8)
40.54 (0.3-1.0)
5+0.22 (0.1-0.4)
Source: Brinton et al (1992)


Women who reported ever having had an induced abortion had the same risk as women reporting never having had an induced abortion (1.00, 0.5-2.0). Women having one or two or more miscarriages had virtually the same risk as women reporting no miscarriages (0.99-1.09). It was concluded that the protective effect was limited to term births.

The epidemiology of endometrial cancer in young women, BE Henderson et al., Br J Cancer 47: 749, 1983.

A case-control study of women, age 45 years or less, at diagnosis in Los Angeles County during 1972-79, found that increasing parity was strongly associated with decreased risk for endometrial cancer. Compared to nulliparous women (1.0), the relative risk was 0.54 for women with one full-term pregnancy, i.e. 28 weeks or more, 0.22 for women with two full-term pregnancies, 0.12 for women with three full-term pregnancies and 0.06 for women with four or more full-term pregnancies. Incomplete pregnancies (spontaneous and induced abortions) were associated with a slight decrease in risk (data not shown). Although the decrease was not statistically significant, it was estimated that 5.6 incomplete pregnancies were equivalent to one full term pregnancy in terms of risk reduction.

A Case-Control Study of Cancer of the Endometrium, JL Kelsey et al., Am J Epidemiology 116(2): 333, 1982.

A study of endometrial cancer in women, aged 47-74, in Connecticut during 1977-79, found that nulliparity and few pregnancies increased the risk of endometrial cancer.

Epidemiology of Endometrial Cancer, M Elwood et al., J National Cancer Institute 59(4): 1055-1060, October 1977.

A case-control study of cancer of the endometrium in the Boston area from 1965-69 found that married women with 3 or more live births had a relative risk of 0.3, married women with one or two live births had a relative risk of 0.6, compared to married nulliparous women (1.0). A history of one or more stillbirths or miscarriages compared to women with no history was 1.1 (0.8-1.6).

1, 2