One approach to the treatment of obstructive azoospermic cases in whom either vasovasostomy cannot be performed or the vas deferens is absent is to construct an artificial spermatocele in the epididymis. By using this method to recover sperms for use in AIH, we have succeeded in achieving pregnancy in only 2 of 35 cases, leading us to investigate the fertility of the recovered sperms. Sperm fertility was investigated by the hypoosmotic swelling (HOS) test, observation by confocal laser-scanning microscope (CLSM), flow cytometry and hamster egg sperm penetration test. The percentage of swollen sperm determined by the HOS test was 15.0-83.3%, with a mean of 47.6 +/- 16.1%, significantly lower than the values of the normal group and the infertile male group, excluding cases of obstructed azoospermia. Examination by confocal laser-scanning microscope of samples stained with PI and FITC-PSA or PI and FITC-Con A stain revealed viable and dead spermatozoa as well as viable and dead acrosome-reacted spermatozoa. In addition, fertility was evaluated from the distribution of spermatozoa in each area from the flow cytometry percentage. This method was shown previously to be useful for the evaluation of fertility as it demonstrated the presence of numerous spermatozoa in the fertile area of cases who did not succeeded in pregnancy. The hamster egg sperm penetration test yielded a fertility rate of 0-25% with a mean of 8.2 +/- 10.0%, which was significantly lower than the value of the normal group, but was not significantly different from the infertile group. The findings of this study indicated that the fertility of epididymal sperms is low, thereby pointing to the need for studies to improve the materials used in the artificial spermatocele as well as the method of sperm recovery. Furthermore, our findings suggest that flow cytometry may be used to select the epididymal sperms with the highest fertility for sperm recovery.