Risk of spontaneous abortion after antibiotic therapy for chronic endometritis before in vitro fertilization and intracytoplasmic sperm injection stimulation

Objective: To investigate whether cured chronic endometritis (CE) from antibiotic treatment would be associated with a higher risk of spontaneous abortion in the following in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI) treatment. Design: Prospective cohort study. Setting: Tertiary reproductive medicine center. Patient(s): Patients with infertility who underwent a routine hysteroscopy underwent an IVF/ICSI stimulation between January 1, 2019, and December 31, 2020. Women with CE (N = 338) underwent antibiotic therapy, and the assisted reproductive outcomes were then compared with women without CE (N = 7,962). Intervention(s): Chronic endometritis was diagnosed through hysteroscopy and confirmed by histology and immunohistochemistry for CD138. Main Outcome Measure(s): Spontaneous abortion rate after the initial embryo transfer. Result(s): A total of 7,218 patients underwent embryo transfer, with 330 in the cured CE group and 6,888 in the non-CE group. Women with cured CE had a higher rate of spontaneous abortion than did those without CE (11.8% vs. 9.2%; crude odds ratio [OR], 1.32 [0.94, 1.86]), and this difference was statistically significant after adjusting for confounding variables (adjusted OR, 1.49 [1.01, 2.19]). The live birth rate was 43.9% in the cured CE group and 50.5% in the non-CE group (crude OR, 0.77 [0.62, 0.96]; adjusted OR, 0.73 [0.59, 0.92]). The incidence of clinical pregnancy did not differ significantly between the 2 groups (56.1% vs. 60.0%; crude OR, 0.85 [0.68, 1.06]; adjusted OR, 0.83 [0.66, 1.03]). Sensitivity analyses stratified by initial fresh- or frozen-thawed embryo transfer cycles resulted in similar results. Conclusion(s): Chronic endometritis cured with antibiotic therapy was associated with an increased risk of spontaneous abortion among women undergoing IVF/ICSI treatment. The interpretation of the findings is limited by a potential confounding bias.