Genetic and environmental influences on endometriosis in Swedish twins

Abstract

Background: Endometriosis is a common gynecological condition with a substantial economic burden on the society. It is known that both genetic and environmental factors are contributing to the phenotypic development of the disease but the mechanisms of their coexistence in the disease process are poorly understood. Endometriosis is assumed to be caused by multiple genes, however, no candidate gene has been found so far to be associated with endometriosis. Further, genome-wide association studies have found several loci to be associated with endometriosis, but the particular genetic variants have not been described yet. Thus, little is known about heritability so far. Furthermore, the possible types of environmental factors are yet to be investigated. Associations between reproductive, lifestyle factors and endometriosis have been investigated, while the significance of these factors in the development of endometriosis is not well established.

Methods: The study population comprised of all female twins in the Swedish twin registry, aged 20–65 years in Study I and II and aged 20–60 years in Study III. Study I estimated the prevalence of endometriosis and the influence of heredity on occurrence of endometriosis by genetic modeling. The associations between reproductive, lifestyle factors and endometriosis were investigated in Study II. In Study III we examined the validity of self-reported endometriosis and endometriosis-related questions with the data on endometriosis in the inpatient registry (IPR).

Results: The prevalence of endometriosis among Swedish twins was estimated to be 4.3%. The heritability of endometriosis was 47% and the remaining effect of 53% was explained by unique environment. A history of late age at menarche and higher parity with two or more children showed an inverse association with the risk to develop endometriosis, while, infertility was strongly associated with endometriosis. Women who used oral contraceptive pills solely for contraception showed no significant association with endometriosis. Body mass index, level of education, coffee consumption, smoking and alcohol consumption did not show any association with endometriosis. Good agreement was found between selfreported endometriosis and data on endometriosis in IPR and the predictive ability of selfreported endometriosis having an endometriosis diagnosis in IPR was increased when there was information about age and infertility.

Conclusions: In conclusion, the collective findings from this thesis suggest a strong genetic influence on phenotypic manifestation of endometriosis. Infertility and endometriosis are strongly associated with each other, however, the causality and whether they have a common genetic origin remains unknown. Self-reported data on endometriosis may be useful in clinical and epidemiological studies.