Effects of localisation of uterine adenomyosis on outcome of in vitro fertilisation/intracytoplasmic sperm injection fresh and frozen-thawed embryo transfer cycles: a multicentre retrospective cohort study

Background Uterine adenomyosis is a benign disease, common among women in their 40 and 50 s, characterised by ectopic endometrial tissue in the uterine myometrial layer. Adenomyosis causes infertility and has a negative effect on the outcomes of in vitro fertilisation (IVF)/intracytoplasmic sperm injection (ICSI) embryo transfer (ET) cycles. It has also been reported to have different characteristics depending on the adenomyotic lesion localisation. The effect of its localisation on IVF/ICSI-ET outcomes is unclear. This study aimed to investigate whether adenomyotic lesion localisation, assessed using magnetic resonance imaging (MRI), was associated with outcomes of IVF/ICSI-ET cycles. Methods This multicentre, joint, retrospective cohort study analysed the medical records of 67 infertile patients with adenomyosis who underwent IVF/ICSI with fresh and frozen-thawed ET at five participating facilities from January 2012 to December 2016 and for whom MRI data were available. Fifteen patients were excluded; therefore, the MRI data of 52 patients were evaluated by two radiologists. We assessed the localisation of and classified adenomyotic lesions into advanced (invades the full thickness of the uterine myometrium), extrinsic (localised on the serosal side), and intrinsic (localised on the endometrial side) subtypes. Results There were 40 advanced, nine extrinsic, and three intrinsic cases, and the outcomes of 100, 27, and nine ET cycles, respectively, were analysed. Pregnancy loss/clinical pregnancy and live birth rates of the advanced, extrinsic, and intrinsic groups were 64 % (16/25) and 9 % (9/100), 33.3 % (3/9) and 22.2 % (6/27), and 50 % (1/2) and 11.1 % (1/9), respectively. A logistic regression analysis adjusted for age, prior miscarriage, and body mass index showed that the extrinsic group had fewer pregnancy losses (odds ratio 0.06; 95 % confidence interval [CI]: 0.00–0.54, p = 0.026) and more live births (odds ratio 6.05; 95 % CI: 1.41–29.65, p = 0.018) than the advanced group. Conclusions Adenomyotic lesions exert different effects on IVF/ICSI-ET outcomes. Thus, MRI assessments of adenomyosis in infertile patients are beneficial. Establishment of treatment plans based on adenomyotic lesion localisation should be considered.