Outcome of ICSI cycles using frozen-thawed surgically obtained spermatozoa in poor responders to ovarian stimulation: cancellation or proceeding to ICSI?

Objective: To investigate the ICSI outcome of the patients who produced four follicles or less after ovarian stimulation using frozen-thawed surgically obtained spermatozoa.

Study Design: The patient cohort of this study was composed of a carefully selected group of 20 men suffering from obstructive and non-obstructive azoospermia and in whom spermatozoa had been seen in their harvested epididymal aspirates and testicular tissues and the cryopreserved specimens had been used for subsequent ICSI cycles. This group of men represent those in whom the female partners produced only four follicles or less after ovarian stimulation.

Results: For the cases of obstructive azoospermia with MESA (n=6) a total of nine cycles was carried out. Four couples went through one cycle, one couple underwent two cycles, one couple accomplished three cycles. Out of the nine cycles, seven went to embryo transfer. No pregnancy occurred in the MESA-ICSI group of patients. The mean+/-S.D. number of oocytes per cycle was 2.556+/-1.236, the mean+/-S.D. number of embryos per transfer was 1.444+/-1.014. Two couples did not have an embryo replacement because of absence of fertilisation. The mean+/-S. D. number of gonadotropin ampoules was 53.88+/-37.30 and the mean+/-S.D. duration of ovarian stimulation was 13.38+/-4.534 days. For the cases of non-obstructive azoospermia with TESE (n=14) a total of 16 cycles was carried out. Thirteen couples went through one cycle, one couple accomplished three cycles. Out of the 16 cycles, 11 cycles went to embryo transfer. One pregnancy occurred in the TESE-ICSI group of patients, which produced live offspring. The mean+/-S.D. number of oocytes per cycle was 3.00+/-1.211, the mean+/-S.D. number of embryos per transfer was 1.313+/-1.195. Five couples did not have an embryo replacement, four of them because of absence of fertilisation and one because of abnormal fertilisation (3 PN). The mean+/-S.D. number of gonadotropin ampoules was 81. 77+/-53.40 and the mean+/-S.D. duration of ovarian stimulation was 16.71+/-3.667 days.

Conclusion: In our opinion, these data show that it is possible to achieve satisfactory fertilisation rates using frozen-thawed epididymal and testicular spermatozoa obtained from men with obstructive or non-obstructive azoospermia, but the limiting factor in reaching the stage of embryo transfer is the poor ovarian response to stimulation induction. It therefore seems preferable to cancel these cycles, in the hope that a better response might be obtained in a subsequent cycle, avoiding in this way financial and emotional implications.