Work

Sperm source and fertilisation after intracytoplasmic sperm injection in male-factor and combined-factor infertility: confounding by indication and unit of analysis in a retrospective single-centre cohort

Year: 2026
Type: preprint
Study design: Observational Study
Abstract: Objective. To examine whether an apparent difference in fertilisation between ejaculated and surgically retrieved spermatozoa at intracytoplasmic sperm injection (ICSI) in routinely collected single-centre data reflects sperm source or is explained by diagnosis, the unit of analysis and small oocyte cohorts. Materials and Methods. Retrospective single-centre cohort (January 2025–June 2026; STROBE/RECORD) of ICSI cycles coded as male-factor or combined-factor infertility, grouped by recorded sperm source (ejaculated vs surgically retrieved). The principal estimate of fertilisation (fertilised/injected oocytes) came from binomial logistic regression with patient-clustered robust variance, adjusted for maternal age and infertility category. The prespecified cycle-level mean difference (patient-level bootstrap), six sensitivity analyses and total fertilisation failure are also reported; pregnancy fields were assessed for completeness only. Results. Of 118 cycles (110 patients), 113 were evaluable: 91 ejaculated and 22 surgically retrieved (19 testicular, 3 presumed epididymal); azoospermia was documented in 14/22 surgical and 2/96 ejaculated cycles. Pooled fertilisation was 70.1% (295/421) vs 60.0% (57/95); adjusted odds ratio 0.74 (95% CI 0.37–1.48), −6.5 percentage points (−21.8 to 8.8). The prespecified cycle-level difference was −17.7 percentage points (−35.6 to 0.0). Total fertilisation failure occurred in 5.5% vs 22.7% of cycles (P = 0.023), all with one to three injected oocytes; with four or more oocytes, fertilisation was 67.9% vs 64.1%. Pregnancy records were incomplete and not compared. Conclusion. The lower fertilisation with surgically retrieved spermatozoa was imprecise, attenuated by adjustment for infertility category and confined to cycles with very few oocytes. Because sperm source was inseparable from azoospermia, these data cannot guide the choice of sperm source; they illustrate how confounding by indication and the cycle-level unit of analysis can inflate apparent differences in routinely collected ICSI data.
Institution: Cairo University
Language: English
Field: Medicine
Open Access status: green
OpenAlex ID: W7236845226
Cites: 46
Cited by:
Related to: 10