Non-obstructive azoospermia (NOA) is a severe form of male infertility. Although cryopreserved testicular sperm are widely used for intracytoplasmic sperm injection (ICSI), a subset of patients experience poor outcomes with a specific frozen-thawed sample. This study aimed to evaluate whether obtaining fresh testicular sperm via repeat mTESE could serve as a rescue strategy for NOA patients with prior poor or failed ICSI outcomes using cryopreserved sperm. This retrospective paired clinical study included 20 men with NOA who underwent microdissection testicular sperm extraction (mTESE) and two ICSI cycles between December 2019 and February 2024 at a tertiary reproductive center. Each patient first underwent ICSI using cryopreserved-thawed testicular sperm, followed by a second ICSI cycle using fresh testicular sperm. Fertilization rate, 2PN rate, number of available embryos, clinical pregnancy rate, and live birth rate were compared between paired cycles using appropriate statistical tests.As a rescue-focused paired design, this cohort was intentionally enriched for patients with poor first-cycle outcomes and therefore does not represent an unselected NOA population. Switching to fresh testicular sperm significantly improved fertilization rate (86.7% vs. 62.7%, P = 0.0045), number of available embryos (3.50 ± 2.24 vs. 1.05 ± 0.60, P = 0.0005), clinical pregnancy rate (55.0% vs. 15.0%, P = 0.021), and live birth rate (45.0% vs. 5.0%, P = 0.008). The 2PN rate was higher but not statistically significant (79.5% vs. 68.5%, P = 0.094). Progressive motility sperm were more frequent in fresh samples (75% vs. 20%, P = 0.0003). No adverse events occurred. In NOA patients with prior poor or failed ICSI outcomes using cryopreserved testicular sperm, repeating mTESE to obtain fresh sperm was associated with significantly improved fertilization and pregnancy outcomes. These findings support fresh testicular sperm as an individualized rescue option rather than evidence of universal superiority over cryopreserved sperm in the general NOA population.
Zhang et al. (Wed,) studied this question.