Effect of uterine dimensions on live birth rates after single embryo transfer in infertile women

Do uterine size parameters measured by baseline transvaginal ultrasound predict live birth after single embryo transfer (SET) of a high-quality blastocyst? Retrospective cohort study including women undergoing their first SET between August 2010 and March 2014 at a large university hospital reproductive centre. The effects of baseline uterine dimensions on live birth rate (LBR) were analysed while controlling for confounding effects. A total of 437 nulliparous and 70 parous women were included. The nulliparous group had lower body mass index (BMI) (24.4 ± 5.1 versus 25.9 ± 4.5 kg/m2; P = 0.015) and a higher number of fibroids (0.4 ± 1.0 versus 0.2 ± 0.5; P = 0.005) than the parous group. While controlling for confounding effects, none of the uterine parameters appeared to be a significant predictor of LBR among nulliparous and parous women (P > 0.05 in all cases). A subsequent analysis of endometrial length was done, whereby the endometrial lengths were divided into quartiles (20.0–32.2 mm; 32.3–36.5 mm; 36.6–40.0 mm; 40.1–54.0 mm). After controlling for confounders, the shortest quartile in the nulliparous group had a significantly lower LBR (P = 0.02) than the other groups. Receiver operating characteristic curves suggested that endometrial cavity length and cervical length did not aid clinically. Uterine parameters do not have a clinically useful impact on LBR after SET of a blastocyst in infertile women. The use of baseline endometrial length to predict live birth is no better than chance, while cervical length only predicts failure to live birth.