Defining associated factors for total fertilization failure and embryo development arrest in intracytoplasmic sperm injection cycles

Total fertilization failure (TFF) and embryo developmental arrest (EDA) remain clinically challenging outcomes in intracytoplasmic sperm injection (ICSI) cycles. Although ICSI is widely and effectively used to overcome fertilization disorders, fertilization failure and embryo development arrest continue to occur in a subset of patients. This study aimed to identify baseline clinical characteristics and ovarian reserve parameters associated with the risk of TFF and EDA. In this retrospective cohort study, 1,846 ICSI cycles performed between January 2016 and December 2019 at a tertiary assisted reproduction center were analyzed. Cycles were categorized as successful fertilization, total fertilization failure (TFF), or embryo developmental arrest (EDA). For regression analyses, TFF and EDA were combined as unsuccessful fertilization. Baseline pre-treatment variables, including female age, basal follicle-stimulating hormone (FSH), serum anti-Müllerian hormone (AMH), antral follicle count, infertility etiology, and semen characteristics, were evaluated. Univariate and multivariable logistic regression analyses were performed, and internal validation was conducted using bootstrap resampling. Successful fertilization occurred in 76.5% of cycles, whereas TFF and EDA were observed in 6.6% and 16.9%, respectively. In univariate analyses, female age, basal FSH, serum AMH, antral follicle count, estradiol level on the day of oocyte pick-up, and oocyte yield parameters were significantly associated with fertilization outcomes. In the multivariable model restricted to baseline variables, female age (adjusted OR 0.94, 95% CI 0.91–0.97), basal FSH (adjusted OR 0.93, 95% CI 0.89–0.97), serum AMH (log-transformed; adjusted OR 1.42, 95% CI 1.18–1.72), and infertility etiology remained independently associated with fertilization success. The optimism-corrected AUC was 0.63. Diminished ovarian reserve markers were independently associated with increased risk of TFF and EDA. Although statistically significant, the predictive performance was modest, supporting use in pre-treatment counseling rather than definitive prediction.