Statistical analysis
Continuous baseline characteristics were represented as mean ± standard deviation (SD) or the median and range, depending on their distribution. Categorical variables were represented as frequency with percentages. The differences about baseline characteristics and pregnancy outcomes between subgroups were compared using the independent-samples t test for continuous variables or Chi-square test for categorical variables. A binary logistic regression was performed to evaluate the associations between LMs with different causes and subsequent pregnancy outcomes, including early miscarriage rate (EMR), late miscarriage rate (LMR), preterm birth rate (PBR) and live birth rate (LBR). Potential confounders were also adjusted, which included maternal age, body mass index (BMI), basal FSH level, number of bilateral basal follicles, PCOS, prior early spontaneous miscarriage, prior live birth, controlled ovarian hyperstimulation (COH) protocol, endometrial thickness on hCG trigger day. LMs, controlled ovarian hyperstimulation (COH) protocol, and endometrial thickness on hCG trigger day were considered as multi-categorical variables, and transformed into dummy variables. The associations were presented as an adjusted OR (aOR) with a 95% CI.
P values were given for two-sided tests and statistical significance was defined by P < 0.05. All of the statistical analyses were performed using SPSS version 26.0 (Chicago, IL, USA).