Semen analysis and fertility prognosis in andrological patients

Semen analyses of 529 men who consulted our department due to infertility problems, were related to the time period prior to conception, with factors adversely affecting the fertility of the female partner taken into consideration. The statistical method used was Cox's proportional-hazard model of regression. Untransformed, logarithmically transformed and dichotomized semen analysis variables were included in the calculations. The relationship between the following parameters and the probability of conception was examined: sperm count, sperm motility, progressive sperm motility, morphology and sperm motility remaining 24 h after ejaculation. All variables co-varied with the probability of conception; however, the exact type of relationship could not be determined by regression analyses. Cox's model assumes an exponential relationship. Our data suggest that this assumption is not suitable for fertility investigations. Using conventionally defined limiting values for normal and pathological semen quality, statistical analysis yielded significant differences in fertility between both categories for all of the variables considered; in the stepwise regression analysis, however, it could be shown that progressive motility and morphology alone were sufficient to discriminate between normal and pathological semen quality. The results are interpreted as indicating that, as a result of semen analysis, it is possible to predict the individual probability of conception if the exact shape of the relationship can be determined, which, up to now, has not been accomplished.