Analysis of intracytoplasmic sperm injection procedures related to delayed insemination and ejaculated, epididymal and testicular spermatozoa.

In all, 1210 treatment cycles were divided into three categories for retrospective analysis according to the period of delay between oocyte retrieval (occurring at a fixed time after human chorionic gonadotrophin) and intracytoplasmic sperm injection (ICSI) of <3 h, 3-5 h, >5 h (referred to as 'delayed ICSI'). Three stages from oocyte to the birth of a live baby were identified for statistical analysis, (i) fertilization (2PN zygotes), (ii) cleavage of 2PN zygotes, (iii) transferred embryo to live birth. Stages 1, 2 and 3 were analysed statistically for the three time periods. Chi-square analysis showed no significant effect of delayed ICSI on fertilization (chi(2) = 3.615, P = 0.65), and embryo transfer to birth (chi(2) = 1.840, P = 0.399). The effect on cleavage was significant (chi(2) = 9.625, P = 0.008). However, shorter incubation times produced results which were better than the traditional longer ones. The success rate at the cleavage stage was so high that the marginal advantage had very little effect on the overall process. This study of a substantial patient sample establishes that ICSI on a peri-ovulatory oocyte (<3 h after oocyte retrieval) does not compromise outcome parameters, and that longer periods of incubation (>5 h) do not offer a statistically significant advantage.