OBJECTIVETo explore various factors affecting the clinical pregnancy outcomes of artificial insemination with donor sperm (AID).

METHODSWe retrospectively analyzed 15,744 cycles of AID in 6302 women and investigated the association of the clinical pregnancy outcomes of AID with the treatment protocols, the times of insemination per cycle, the age of the infertile women, the status of the oviduct, and the number of AID cycles.

RESULTSThe pregnancy rate of AID was higher in the chlomiphene-treated women than in those of the natural cycle group (P = 0.003) but showed no significant differences either between the chloramiphene and human menopause gonadotropin (HMG) or between the HMG and natural cycle groups (P > 0.05), and so was it in the women that had received AID twice per cycle before and after ovulation (26.3%) than in those that had undergone only once before (7.0%) or after ovulation (23.7%) (P < 0.05). However, the pregnancy rate was remarkably lower in the women aged 35-40 years (16.5%), especially in those over 40 years (1.2%), than in those under 35 years (26.0%) (P < 0.05). There was no significant difference in the success rate of AID between the women with oviductal adhesion and those without (27.4% vs. 28.1%, P > 0.05). The pregnancy rate of the first cycle of AID (27.6%) was markedly higher than those of the second (24.7%), third (23.9%), and fourth (23.1%) (P < 0.01), but with no significant differences among the latter three cycles (P > 0.05), while that of the fifth cycle (19.0%) was remarkably lower than those of the first four (P < 0.01).

CONCLUSIONThe age of the infertile women is an important factor affecting the success rate of AID. AID twice per cycle is better than once only. For those without oviductal factors, at least 4 cycles of AID are required before in vitro fertilization.