LAUSR

What live birth rate do we see when we use testicular sperm in ART for non-azoospermic couples after at least one previous failed cycle? In our cohort of couples 24% had a live birth using testicular sperm and therefore was not higher than national average ART rates. There is increased interest in using testicular sperm in assisted reproduction technology (ART) to improve outcomes after previous failed cycles. Mehta et al. reported results of a 50% live birth rate using testicular sperm in the first cycle for couples with oligospermia and a history of failed cycles with ejaculated sperm. We aim to audit our results in a similar population of couples. St Peters Andrology Centre in London, United Kingdom completed 128 surgical testicular sperm retrievals reviewed between the two-year period of 2018–2019. We conducted a retrospective audit of their paper-based records to identify those couples with injectable sperm on their semen analysis and who had previous cycles attempts using ejaculated sperm. We identified 27 couples who underwent testicular sperm extraction despite having an ejaculated semen analysis with injectable sperm and at least one previous failed cycle. A systematic review of their paper and electronic medical record was conducted to assess live birth rates and fertilization rates from ART. Couples had an average male age of 41 (range 31–60) and an average female age of 38 (range 30–45). The men had an average serum testosterone of 15 nmol/L (range 8–35 nmol/L) and an average serum FSH of 8.9 IU/L (range 1.7–30 IU/L). 59% (n = 17) of men had a DNA fragmentation index completed with an average score of 41% (range 31%–51[Y1]%). In the women the mean serum anti-Müllerian hormone (AMH) was 15.8 pmol/l (range 1–64 pmol/l). With ejaculated sperm the fertilization rate was 59% (95% CI [27%, 59%]) and blastocyst conversion rate was 43% (95% CI [50%, 69%]). There was no statistical significance with testicular sperm where the fertilization rate was 58% (95% CI [51%, 65%]) and blastocyst conversion rate was 54% (95% CI [40%, 67%]). Overall, there were 7 clinical pregnancies in this population of couples. Of these clinical pregnancies, 2 miscarried and 5 progressed to a live birth. This audit yielded a live birth rate per cycle of 15% and a live birth rate per couple of 24%. Limitations of the study are low number of patients and absence of a control group. Wider implications of the findings: We recommend caution and further analysis going forward using testicular sperm in ART where ejaculated sperm in available. Not applicable