Prognostic variables in predicting pregnancy. A prospective follow up study of 907 couples with an infertility problem.

This report describes a prospective follow-up study on 907 infertile men presenting as partners of infertile couples. Couples who conceived were compared to those who remained infertile during a 3-year follow up. A univariate life table analysis showed that the duration of infertility, the age of both spouses, a history of male urethritis, the body mass index, the percentage of motile spermatozoa, the quality of motility, the motile sperm density, the total motile count, the sperm morphology, and the pH of semen were significant predictors of fertility. Another eleven socio-demographic and nine semen variables were not associated with fertility. Cox multiple regression analysis identified the duration of infertility, percentage total motility, the quality of motility combined, a history of male urethritis, the wife's age, and the morphology of the spermatozoa to be independent predictors of the reproductive outcome. A prognostic 'risk' ratio (chance for achieving pregnancy) was calculated as a function of these variables for each couple and a life table analysis was made between three classified subgroups of couples with different risk ratios. These life table curves make it possible to define the probability of an individual couple to achieve pregnancy at any given time.