Purpose
Little is known about the effectiveness and safety of varicocele repair in patients with male factor infertility undergoing ICSI treatment. Data concerning neonatal outcomes is lacking. We aimed to investigate the effect of varicocele repair prior ICSI.

Methods
We analyzed retrospective data from June 2016 to January 2024 using propensity score regression analyses. We compared embryological, clinical and neonatal outcomes. We also evaluated the effect of cofounders on the rate of live birth using multivariate analysis.

Results
Data from 582 ICSI cycles showed no association between livebirth and varicocele repair for male-factor patients undergoing ICSI (aOR 0.81[95% CI0.52, 1.25]; p = 0.3) in the adjusted multilevel logistic analysis. Four hundred and four ICSI cycles were included in the propensity score regression analysis, which confirmed the same non-significant increase of livebirth and other embryological and clinical outcomes. Data about neonatal outcomes were similar between the two groups, except for the rate of very preterm birth were higher in the varicocelectomy group compared to the untreated varicocele group.

Conclusion
In male-factor infertility associated with varicocele and undergoing ICSI treatment, varicocelectomy does not significantly increase the odds of live birth compared with untreated varicocele. The increased number of very preterm birth in the varicocelectomy group may be related to the high number of multiple gestations exists in the varicocelectomy group and may not correlate to the procedures itself. Further well-powered prospective trails remain warranted.