Salpingectomy for ectopic pregnancy reduces ovarian cancer risk—a nationwide study

Abstract

Recent studies propose fallopian tubes as the tissue origin for many ovarian epithelial cancers. To further support this paradigm, we assessed whether salpingectomy for treating ectopic pregnancy had a protective effect using the Taiwan Longitudinal National Health Research Database. We identified 316 882 women with surgical treatment for ectopic pregnancy and 3 168 820 age- and index-date-matched controls from 2000 to 2016. In a nested cohort, 91.5% of cases underwent unilateral salpingectomy, suggesting that most surgically managed patients have salpingectomy. Over a follow-up period of 17 years, the ovarian carcinoma incidence was 0.0069 (95% confidence interval [CI] = 0.0060 to 0.0079) and 0.0089 (95% CI = 0.0086 to 0.0092) in the ectopic pregnancy and the control groups, respectively (P < .001). After adjusting the events to per 100 person-years, the hazard ratio (HR) in the ectopic pregnancy group was 0.70 (95% CI = 0.61 to 0.80). The risk reduction occurred only in epithelial ovarian cancer (HR = 0.73, 95% CI = 0.63 to 0.86) and not in non-epithelial subtypes. These findings show a decrease in ovarian carcinoma incidence after salpingectomy for treating ectopic pregnancy.

Ovarian cancer is one of the most aggressive types of women’s malignancy (1). It can be broadly divided into two major types: epithelial cancer (carcinoma) and nonepithelial cancer (2). Risk factors associated with epithelial ovarian cancer include germline mutations in genes involved in homologous recombination DNA damage repair (mainly BRCA1, BRCA2, RAD51C, RAD51D, and PABL2) and family history of breast and ovarian cancer (1,3,4). Epidemiologic studies have also shown the number of lifetime ovulatory years to be positively associated with risk, whereas oral contraceptive use and increased parity are inversely associated with risk (1,5,6).

Ovarian epithelial cancer is unique in that most ovarian carcinomas do not derive from their “cognate” organs (ie, the ovaries). First proposed 2 decades ago, the fallopian tube paradigm of ovarian carcinoma upends previous concepts that ovarian carcinoma is of ovarian origin (7-20). The current paradigm posits that fallopian tube epithelial cells are the origin of most high-grade serous ovarian carcinomas, and the cells from the fallopian tube precursor lesion, serous tubal intraepithelial carcinoma (STIC), exfoliate onto the ovaries, omentum, and peritoneal surfaces to give rise to advanced stages of the disease (7).

Because opportunistic salpingectomy, the practice of removing both fallopian tubes for ovarian cancer prevention at the time of another postreproductive intra-peritoneal surgery, has not yet been universally adopted in North America, directly assessing the impact of salpingectomy on ovarian cancer incidence from archived databases is generally not feasible. Here, we undertook a nationwide population-based study in Taiwan to determine whether salpingectomy for the surgical management of a benign condition affected ovarian cancer risk. We focused on ectopic pregnancy because the majority (>95%) of ectopic pregnancies occur in fallopian tubes, and ectopic pregnancy is one of the most common indications for salpingectomy in reproductive-age women (21).

Data were extracted from the Taiwan Longitudinal National Health Insurance Research Database (NHIRD). NHIRD was launched in 1996 and includes data accumulated from a single-payer National Health Insurance program since 1994, covering nearly 99% of the entire Taiwanese population (23 million as of April 2023). NHIRD contains deidentified data, including disease codes, procedure codes, and demographic information from outpatient visits, emergency care, and inpatient services (22,23).

On the basis of the diagnostic code for ectopic pregnancy (ICD-9-CM633), we identified 316 882 women undergoing emergency surgery for ectopic pregnancy from 2000 to 2016. We used Taiwan’s cancer registry to track the patients with ovarian cancer up to 2017. The cases were grouped into three age tiers: younger than 31, 31-40, and 41 years or older (Table 1). The controls consisted of cancer-free women who, at the time of the index case, had no record of ectopic pregnancy. Controls were randomly selected after matching for age and index dates. The case to control ratio was 1 to 10 (Table 1).

During the 17-year follow-up period (2000-2016), the ovarian carcinoma incidence was 0.0069 (95% confidence interval [CI] = 0.0060 to 0.0079) in the ectopic pregnancy group and 0.0089 (95% CI = 0.0086 to 0.0092) in the control group (P < .001, log-rank test). After adjusting the events to per 100 person-years, the hazard ratio (HR) was 0.70 (95% CI = 0.61 to 0.80) in the ectopic pregnancy group relative to the control group (Table 1). Figure 1 shows the cumulative risk by years in both case and control groups. Ovarian cancer risk was significantly lower in the ectopic pregnancy compared with the control group over the entire monitoring period (P < .001).

Across the 3 age subgroups, significant differences in risk reduction were observed between cases and controls for women younger than age 31 (HR = 1.35) and aged 31-40 (HR = 1.32) (Table 2). No risk reduction was found for the aged 41 or older subgroup (HR = 1.03). Therefore, the protective benefit of salpingectomy performed in women 40 years old or older was less than those younger than 40 years old. This finding may be attributable to the much smaller case number (15 990) in the oldest subgroup than in the younger subgroups (170 356 and 130 536).

Further analysis showed the risk reduction only for epithelial ovarian cancer (HR = 0.73, 95% CI = 0.63 to 0.86) and not for nonepithelial histological subtypes (HR = 1.16, 95% CI = 0.85 to 1.58) (Table 3). In the epithelial type, the incidence per 100 person-year is 0.003, 0.008, and 0.015 for ages <31, 31-40, and ≥41 groups, respectively. In the nonepithelial type, it is 0.001, 0.0008, and 0.0006 for ages <30, 30-39, and ≥40 groups, respectively. To determine the percentage of treated women undergoing salpingectomy, we analyzed the percentage of this surgical procedure in a nested cohort from 3 hospitals in the same cohort (n = 1596). Analysis of this hospital-based dataset disclosed that 91.5% of ectopic pregnancy patients underwent unilateral salpingectomy.

At least 2 population-wide studies have reported a reduction in ovarian cancer incidence among women who had salpingectomy for various reasons. In a Swedish study (24), ovarian cancer risk decreased by 29% (HR = 0.71) in the unilateral salpingectomy group and 65% (HR = 0.35) in the bilateral salpingectomy group (24). Our results showing a hazard ratio of 0.70 for women with surgically managed ectopic pregnancy is consistent with this Swedish report. Notably, our study population is mainly Taiwanese of Han Chinese and Austronesian descent; therefore, the protective effect after salpingectomy is unlikely to be dependent on ethnic background.

Another observational study from the Canadian Province of British Columbia with a population of 5 million showed that opportunistic bilateral salpingectomy during hysterectomy for benign indications or instead of bilateral tubal ligation reduces epithelial ovarian cancer risk (25). In that report, there were no serous carcinomas documented in 25 889 women undergoing opportunistic salpingectomy, whereas the age-adjusted expected number was 8.68 epithelial ovarian cancers.

Our finding of risk reduction only to the epithelial type of ovarian cancer is in line with the fallopian tube as the tissue of origin for many epithelial ovarian cancers because nonepithelial ovarian cancers originate from ovarian stroma or germ cells. The tubal paradigm in ovarian cancer genesis has several important implications not only for studying the biology of ovarian cancer but also for the development of early detection strategies and the clinical management of ovarian cancer patients (7). This model explains why most high-grade serous carcinomas are widely metastatic at first diagnosis: because the precursor lesions—STIC from the fallopian tube mucosa—readily disseminate onto the peritoneal surfaces of the abdominal cavity. On the other hand, there is potential for cancer or even STIC cells to travel retrograde through the uterine cavity and to the cervix, where a routine liquid-based Pap smear might collect them for early detection using sensitive and specific molecular tests (26). The paradigm provides a biological foundation for adopting opportunistic salpingectomy for the primary prevention of ovarian carcinoma in average-risk women (women who have no known increased risk because of genetic or familial factors) when there is an opportunity to remove fallopian tubes during another intra-abdominal surgery (25,27-29).

Endometriosis has been proposed as a causative factor in ectopic pregnancy (30,31) and is a risk associated with ovarian epithelial cancer (32-35). In this cohort, we also observed an independent association between endometriosis and ovarian cancer risk with an adjusted hazard ratio of 3.19 (95% CI = 2.96 to 3.43), consistent with a recent comprehensive meta-analysis that revealed a positive association between ovarian cancer risk and endometriosis (32). We then compared the numbers of women with endometriosis between ectopic pregnancy and control groups (without ectopic pregnancy) in our NHIRD cohort. There were 63 605/316 882 (20.1%) diagnoses of endometriosis in the ectopic pregnancy group vs 435 242/3 168 820 (13.7%) in the control group. We found endometriosis to be a relatively weak risk factor for ectopic pregnancy in the NHIRD cohort with a hazard ratio of 1.58 (95% CI = 1.56 to 1.59), consistent with a prior report that was based on the Taiwanese population (31). Given the above observations, we would have expected to find a higher incidence of ovarian cancer in the ectopic pregnancy group, as this group had a higher number of women with a history of endometriosis. However, our outcome data showed the opposite effect. Thus, salpingectomy is the most reasonable explanation for the ovarian cancer risk reduction observed among women with ectopic pregnancy in our study.

In summary, our analysis of NHIRD provides evidence that surgical management of ectopic pregnancy confers protection against the development of epithelial ovarian cancer. We analyzed a nationwide population-based dataset encompassing 23 million residents, so selection bias is minimal. However, our investigation has limitations. Foremost, because the NHIRD is for reimbursement purposes, we cannot rule out coding inaccuracy in some of the data. A future centralized pathology review is needed to verify ovarian cancer histologic subtypes and to determine their individual impact on risk reduction. The current study did not address the protective effects of surgical vs medical treatment of ectopic pregnancy or the dose-dependent effects of a second salpingectomy. Nevertheless, our results support the tubal paradigm of ovarian cancer development and the increasing advocacy of salpingectomy as a primary prevention strategy for ovarian cancer in average-risk women.

Data availability

The deidentified NHIRD data are available for researchers to access following the application and authorization by the Department of Health and Welfare Data Center in Taiwan (23).

Author contributions

Ju-Chuan Yen, MD (Conceptualization; Formal analysis; Investigation; Methodology; Writing—original draft), Tzu-I Wu, MD (Conceptualization; Investigation), Rebecca Stone, MD (Conceptualization; Writing—original draft; Writing—review & editing), Tian-Li Wang, PhD (Conceptualization; Resources; Writing—original draft; Writing—review & editing), Kala Visvanathan, MD (Conceptualization; Investigation; Methodology; Writing—review & editing), Li-Ying Chen, BS (Data curation; Resources; Validation), Min-Huei Hsu, MD, PhD (Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Resources; Supervision; Validation), Ie-Ming Shih, MD, PhD (Conceptualization; Funding acquisition; Investigation; Writing—original draft; Writing—review & editing).

Funding

This study is supported by the Richard W. TeLinde Endowment, the Johns Hopkins University (to IMS), Break Through Cancer-IOC (to RS and IMS), NIH/NCI P50 CA228991 (to IMS and TLW), and NSTC112-2321-B-038-005 (to MHH).

Conflicts of interest

None of the authors claims any conflict of interest in this study.

Acknowledgments

We acknowledge the above funding for the salary support of some authors who were involved in the design, analysis, and manuscript preparation of this study. In addition, the NSTC grant supports the collection, analysis, and interpretation of the data.

References

Author notes