A Sonographic Endometrial Thickness <7mm in Women Undergoing In-Vitro Fertilization Increases the Risk of Placenta Accreta Spectrum
Siying Lai,Lizi Zhang,Yang Luo,Zhongjia Gu,Zhenping Yan,Yuliang Zhang,Yingyu Liang,Minshan Huang,Jingying Liang,Shifeng Gu,Jingsi Chen,Lei Li,Dunjin Chen,Lili Du
DOI: https://doi.org/10.1016/j.ajog.2024.02.301
IF: 9.8
2024-01-01
American Journal of Obstetrics and Gynecology
Abstract:Background The prevalence of placenta accreta spectrum (PAS), a potentially life-threatening condition, has exhibited a significant global rise in recent decades. Effective screening methods and early identification strategies for PAS could enable early treatment and improved outcomes. Endometrial thickness (EMT) plays a crucial role in successful embryo implantation and favorable pregnancy outcomes. Extensive research has been conducted on the impact of EMT on assisted reproductive technology cycles, specifically in terms of pregnancy rates, live birth rates, and pregnancy loss rates. However, limited knowledge exists regarding the influence of EMT on PAS. Objective This study aims to evaluate association between pre-implantation EMT and the occurrence of PAS in women undergoing assisted reproductive technology cycles. Study Design A total of 4637 women who had not undergone prior caesarean section, conceived by in vitro fertilization or intracytoplasmic sperm injection embryo transfer treatment and subsequently delivered at the Third Affiliated Hospital of Guangzhou Medical University between January 2008 to December 2020 were included in this study. To explore the relationship between EMT and PAS, we employed smooth curve fitting, threshold effect, and saturation effect analysis. Multivariate logistic regression analysis was performed to evaluate the independent association between EMT and PAS while adjusting for potential confounding factors. Propensity score matching (PSM) was performed to reduce the influence of bias and unmeasured confounders. Furthermore, we employed causal mediation effect analysis to investigate the mediating role of EMT in the relationship between gravidity and ovarian stimulation protocol and the occurrence of PAS. Results Among the 4637 women included in this study, pregnancies with PAS (159, 3.4%) had significantly thinner EMT (non-PAS 10.08±2.04 vs PAS 8.88±2.21 mm，P<0.001) during the last ultrasound before embryo transfer. By smooth curve fitting, the saturated effect was found on EMT and PAS, with an inflection point at 10.9mm. Then the EMT was divided into four groups: ≤7mm, >7 to≤10.9mm, >10.9 to≤13mm, >13mm, and the absolute rates of PAS in each group was 11.91%, 3.73%, 1.35%, and 2.54%, respectively. When compared to women with an EMT ranging from between 10.9-13mm, the odds of PAS increased from adjusted odds ratio (aOR) of 2.27 (95% CI, 1.33-3.86) for EMT ranging from 7-10.9mm to aOR of 7.15 (95% CI, 3.73-13.71) for EMT <7mm after adjusting for potential confounding factors. Placenta previa remained as an independent risk factor for PAS (aOR, 11.80; 95% CI, 7.65-18.19). Moreover, EMT <7mm was still an independent risk factor for PAS (aOR, 3.91; 95% CI, 1.57-9.73) in the matched cohort after PSM. Causal mediation analysis revealed that approximately 63.9% of the total effect of gravidity and 18.6% of the total effect of ovarian stimulation protocol on PAS were mediated by the EMT. Conclusions The findings from our study indicate that thin EMT is an independent risk factor for PAS in women without prior caesarean section undergoing assisted reproductive technology treatment. The clinical significance of this risk factor is slightly lower than placenta previa. Furthermore, our results demonstrate that EMT plays a significant mediating role in the relationship between gravida or ovarian stimulation protocol and PAS.