First-line surgery versus first-line assisted reproductive technology for women with deep infiltrating endometriosis: a systematic review and meta-analysis.

BACKGROUND: The efficiency of different first-line treatments, such as first-line surgery and assisted reproductive technology (ART), in women with deep infiltrating endometriosis (DIE) is still unclear due to a lack of direct comparative trials. This systematic review and meta-analysis aim to elucidate and compare the efficacies of first-line treatments in patients with DIE, with an emphasis on fertility outcomes.

METHODS: An exhaustive search of PubMed Central, SCOPUS, EMBASE, MEDLINE, Cochrane trial registry, Google Scholar, and Clinicaltrials.gov databases was done to identify studies directly comparing first-line surgery and assisted reproductive technology (ART) for DIE, and reporting fertility-related outcomes. Pooled estimates for each of the binary outcomes were reported as odds ratios (ORs) with 95% confidence intervals (CIs). The results were pooled using a random-effects model with the Mantel-Haenszel technique.

RESULTS: Our results show that pregnancy rate per patient (OR, 1.47; 95% CI, 0.59 to 3.63), pregnancy rate per cycle (OR, 1.16; 95% CI, 0.45 to 2.99), and live births per patient (OR, 1.66; 95% CI, 0.56 to 4.91) were comparable in DIE patients, treated with surgery or ART as a first line of treatment. When both complete and incomplete surgical DIE excision procedures were taken into account, surgery was associated with a significant enhancement in the pregnancy rate per patient (OR, 1.63; 95% CI, 1.11 to 2.40).

CONCLUSION: The available evidence suggests that both first-line surgery and ART can be effective DIE treatments with similar fertility outcomes. However, further analysis reveals that excluding studies involving endometriomas significantly alters the understanding of treatment efficacy between surgery and ART for DIE-associated infertility.

SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=426061, identifier CRD42023426061.