A model for estimating the potency and survival of human gametes in vivo.

Sperm and ova are sensitive to numerous toxicants in animal studies; however, human vulnerability has been far more difficult to assess, due in part to a lack of methods for measuring the viable survival of human gametes in vivo. We present a parametric model for fertility, which assumes that the viable lifetime of the ovum is fixed while that of sperm is exponentially distributed. By reducing the number of parameters that must be estimated, compared to a previous approach, the model leads to improved tests for differences in sperm and egg survival between exposed and unexposed couples. Since it assumes that batches of sperm introduced on different days present independent competing "risks" (of fertilization) to the ovum, the model also provides for estimation of the age distribution, in days, of the sperm which actually fertilized the ova. This allows us to consider whether older sperm are more likely to produce defective embryos. We apply this model to data from a group of women who were intensively studied, beginning when they discontinued contraception in order to start a pregnancy. Participants kept daily records of intercourse. Daily urine specimens allowed the day of ovulation to be estimated and conceptions to be identified, based on assays of excreted hormones. Applying the parametric model to these data, the estimated mean viable lifetime for sperm is 1.4 days, while the lifetime of the ovum appears to be less than a day.(ABSTRACT TRUNCATED AT 250 WORDS)