Endometriosis and Ectopic Pregnancy: A Meta-analysis

Abstract

Objective: To systematically review and perform a meta-analysis of the risk of ectopic pregnancy in endometriosis.

Data sources: MEDLINE (OVID), Embase (OVID), CINAHL (EBSCO), and Cochrane Library to April 1, 2019. Inclusion criteria were cohort or case-control studies from 1990 onward. Exclusion criteria were cohort studies without controls, case reports or series, or no English full-text.

Methods of study selection: A total of 1361 titles/abstracts were screened after removal of duplicates, 39 full-texts were requested, and, after 24 studies were excluded, there were 15 studies in the meta-analysis.

Tabulation, integration, and results: Data were extracted using standardized spreadsheets with 2 independent reviewers, and conflicts were resolved by a third reviewer. We performed random effects calculation of weighted estimated average odds ratio (OR). Heterogeneity and publication bias were assessed with the I² metric and funnel plots/Egger's test, respectively. The Ottawa-Newcastle Quality Assessment Scale was used with a cutoff of ≥7 for higher quality. There were 10 case-control studies (17 972 ectopic pregnancy cases and 485 266 nonectopic pregnancy controls) and 5 cohort studies (30 609 women with endometriosis and 107 321 women without endometriosis). For case-control studies, endometriosis was associated with increased risk of ectopic pregnancy with an OR of 2.66 (95% confidence interval [CI] = 1.14-6.21, p = .02). For cohort studies, the OR was 0.95 (95% CI = 0.29-3.11, p = .94), but after post hoc analysis of the studies with a Ottawa-Newcastle score ≥7, the OR was 2.16 (95% CI = 1.67-2.79, p <.001). For both case-control and cohort studies, there was high heterogeneity among studies (I² = 93.9% and I² = 96.6%, Q test p <.001) but no obvious evidence of systematic bias in the funnel plot, and Egger's test results were not significant (p = .35, p = .70), suggesting no strong publication bias. There were insufficient data to make any conclusions with respect to anatomic characteristics of endometriosis (e.g., stage) or mode of conception (e.g., assisted reproductive technology vs spontaneous).

Conclusion: Possible evidence of an association between endometriosis and ectopic pregnancy was observed (OR = 2.16-2.66). However, these results should be considered with caution, owing to high heterogeneity among studies. Continued research is needed to delineate the pregnancy implications of endometriosis.

Keywords: Ectopic pregnancy; Endometriosis; Meta-analysis; Systematic review.