ACROSOME REACTION INDUCIBILITY PREDICTS FERTILIZATION SUCCESS AT IN-VITRO FERTILIZATION

We prospectively studied the ability of acrosome reaction (AR) inducibility to predict fertilization success in a group of 232 infertile patients presenting sequentially for in-vitro fertilization (IVF). The median percentage of eggs fertilized for the overall patient population was 25% (interquartile range 5-58%), with one to 29 oocytes available for insemination (median, five oocytes). The median percentage of eggs fertilized at IVF increased as the percentage of spermatozoa able to undergo AR became greater: spermatozoa with a faded AR (less than or equal to 5%) fertilized only 12% of eggs, while spermatozoa with AR values > 9% fertilized 50% of eggs. The assay had a specificity of 0.75, a sensitivity of 0.55 and an odds ratio of 2.9; thus, AR-positive patients are 2.9 times more likely to achieve fertilization than patients with a faded AR. Receiver operator characteristic (ROC) curves were constructed for AR, sperm concentration and percentage of normal forms in semen. All three parameters proved to be potentially useful in predicting the occurrence of fertilization, although AR and morphology appeared to be better than sperm concentration by ROC analysis. Patients were divided into four clearly defined subgroups according to their traditional semen characteristics, including morphology. The median percentage of eggs fertilized decreased as traditional semen characteristics deteriorated, from a median of 46% for patients with excellent sperm concentration, motility and morphology, to a median of 29% for patients with suboptimal semen quality and a median of 0% for patients with severely impaired semen. Within each patient subgroup, the median percentage of eggs fertilized was 3- to 4-fold higher for individuals with a positive AR than for those with a failed AR, indicating that AR has a greater effect on fertilization rate than traditional semen parameters including morphology. We now recognize that some men with good semen characteristics have an unexpectedly poor AR and a markedly reduced fertilization rate, while other men with poor traditional semen characteristics unexpectedly retain AR and perform relatively well at IVF. By contrast to AR, morphology seemed to have little effect on fertilization success (two-way analysis of variance not significant). The wife's age and oocyte quality were evenly distributed among the different patient subgroups, indicating that differences in fertilization rate could not be attributed to either parameter. Our data indicate that AR has a much higher predictive value for IVF success than traditional semen parameters including morphology. We propose that AR assessment is a clinically useful diagnostic tool in determining a patient's likelihood of achieving fertilization at IVF.