Use of testicular versus ejaculated sperm for intracytoplasmic sperm injection among men with cryptozoospermia A meta analysis

Among cryptozoospermic men, use of testicular sperm for intracytoplasmic sperm injection offers no advantages in fertilization or pregnancy rates when compared with ejaculated sperm.

Nikita Abhyankar, M.B.ChB., Martin Kathrins, M.D., Craig Niederberger, M.D.

Volume 105, Issue 6, Pages 1469-1475


To examine outcomes of intracytoplasmic sperm injection (ICSI) using testicular versus ejaculated sperm among men with cryptozoospermia.


Not applicable.

Men with cryptozoospermia undergoing consecutive ICSI cycles using ejaculated or testicular sperm.

A systematic search was performed using PubMed (inception to August 2015). Inclusion criteria were studies comparing ICSI outcomes among men with cryptozoospermia using ejaculated and testicular sperm.

Main Outcome Measure(s):
Primary outcomes included ICSI fertilization or pregnancy rates (PRs). Secondary analysis included number of retrieved oocytes, maternal and paternal ages. Meta-analysis of weighted data using a random effects model was performed. Results are reported as relative risk or weighted mean differences (WMD) with 95% confidence intervals (CI).

Five cohort studies were included, encompassing 272 ICSI cycles and 4,596 injected oocytes. There were no differences in ICSI PRs (relative risk [RR] 0.53, 95% CI 0.19–1.42, I2 = 67%) or fertilization rates (RR 0.91, 95% CI 0.78–1.06, I2 = 73%) between testicular and ejaculated sperm groups. There was a significant trend toward increasing maternal age (WMD 1.69 years, 95% CI −2.71 to −0.66) and paternal age (WMD 2.61 years, 95% CI −4.73 to −0.48) with testicular sperm. There was no difference between numbers of oocytes retrieved (WMD 0.95, 95% CI −0.15 to 2.05). Post-hoc power analysis revealed pβ <20% for PR analysis and pβ <10% for fertilization rate analysis. Conclusion(s):
The existing literature does not support a recommendation for men with cryptozoospermia to use testicular sperm in preference over ejaculated sperm for ICSI.

  • Nikita Abhyankar

    Many thanks for the thoughtful comments. Unfortunately, complete data on DNA fragmentation index was not available and therefore not included in this meta-analysis. Part of our initial research question was whether fertilization and pregnancy outcomes would differ because of the two sperm sources having potentially differing DFIs. However, we were not able to demonstrate that this was case. As Dr Ko identified, the maternal and paternal age differences may be a confounding factor. Whether, on an individual level DNA fragmentation and sperm source makes a difference, we were unable to answer from our study.

  • Interesting paper that may impact the recommendation for couples that have failed IVF/ICSI with ejaculated sperm to try a cycle with testicular sperm. We have evidence sperm extracted from the testicle directly has lower DFI, but it appears in the current meta analysis of the current literature that the lower DFI does not improve fertilization rates. In the studies reviewed, however, the maternal and paternal ages could be factors impacting outcomes. Will be interesting to see if this recommendation will change as more data is obtained.

  • msamplaski

    I completely agree with Jason, and these are two times that this question comes up routinely. We have accumulating evidence that TESE sperm have lower DNA frag rates in men with elevated DNA frag rates. And in couples who have failed IVF once, they are often looking to modify factors so that subsequent IVF cycles are successful; The issue of TESE sperm is often one of the questions that is asked.

  • Jason Kovac

    How about the effects of using testicular sperm on couples who have previously utilized ejaculated sperm for ICSI and failed. Any thoughts about examining the role of DNA fragmentation in the meta-analysis? Was that examined? Probably not enough data at the present time.

Translate »