Endometriosis and the subsequent risk of epithelial ovarian cancer.

OBJECTIVE

There is a possible correlation between endometriosis and an increased risk of epithelial ovarian cancer (EOC), but many uncertainties remain, including race, exposure or surveillance time, and surgical confirmation. Therefore, we carried out a large-scale, nationwide, controlled cohort study in the Taiwanese women to respond to these uncertainties.

MATERIALS AND METHODS

A historical cohort study was performed by linking the National Health Insurance Research Database of Taiwan. Each patient diagnosed with endometriosis (n = 7537) between 2000 and 2009 was background matched with up to two women without endometriosis (n = 15,074). The total was 136,643 person-years of follow-up and 24 women having new EOC. Cox regression analysis was used to determine the relationship between the EOC incidence rate and an endometriosis status.

RESULTS

The EOC incidence rate of the endometriosis and non-endometriosis women was 3.31 per 10,000 person-years and 0.99 per 10,000 person-years, respectively, contributing to an adjusted hazard ratio (HR) of 3.28 (95% confidence interval, 1.37-7.85). The women with surgical confirmation had a much higher adjusted HR (3.87; 95% confidence interval, 1.58-9.47). No significantly statistical difference of surveillance time between women with and without endometriosis (3.87 years vs. 3.73 years). The occurrence of EOC was not also affected by exposure time of women with endometriosis.

CONCLUSION

Taiwanese women with endometriosis really had a risk of newly developed EOC, especially those who had a surgical diagnosis, and this three-fold increase of risk was neither influenced by exposure time nor biased by surveillance.