The presence of excess leucocytes in the semen has been associated with male infertility. According to the WHO, concentrations of more than 106 leucocytes ml-1 are considered as leucocytospermia, indicating genital tract infections. Up to now, no consensus has been achieved on how leucocytes should be quantified in semen. Using the peroxidase staining and monoclonal antibodies to CD15, CD45 and CD68, we found significant differences between the detection methods. Only 47.4% of the semen samples that were assessed as leucocytospermic by CD45 were identified as such by peroxidase staining. The concentration of peroxidase-positive cells was significantly correlated with polymorphonuclear granulocyte (PMN) elastase (P < 0.0001). However, a negative correlation of peroxidase-positive cells with the sperm concentration was only found in oligozoospermic patients (P < 0.0001). Moreover, the slightly positive correlation with normal sperm morphology seems to be applicable only in cases of oligozoospermia. Significant negative correlation of the number of peroxidase-positive cells were found for both maximal inducible acrosome reaction (P = 0.0219) and the inducibility of acrosome reaction (P = 0.0370), indicating a rather deleterious effect of leucocytes on this important sperm function. Concerning the result in the in vitro fertilization programme, none of the examined parameters (PMN elastase, concentration of round cells and peroxidase-positive cells) showed a correlation with either fertilization or pregnancy. This result seems to be reasonable as severely damaged spermatozoa and leucocytes are eliminated from the ejaculate by different sperm separation methods. Interestingly, a significant negative correlation of the TUNEL assay as a measure of sperm DNA fragmentation was found only with pregnancy (P = 0.006) but not with fertilization. As DNA fragmentation can also be caused by ROS that are generated by leucocytes, this causality should not be neglected.

The evaluation of different functional sperm parameters has become a tool in andrological diagnosis. These assays determine the sperm's capability to fertilize an oocyte. It also appears that sperm functions and semen parameters are interrelated and interdependent. Therefore, the question arose whether a given laboratory test or a battery of tests can predict the outcome in in vitro fertilization (IVF).One-hundred and sixty-one patients who underwent an IVF treatment were selected from a database of 4178 patients who had been examined for male infertility 3 months before or after IVF. Sperm concentration, motility, acrosin activity, acrosome reaction, sperm morphology, maternal age, number of transferred embryos, embryo score, fertilization rate and pregnancy rate were determined. In addition, logistic regression models to describe fertilization rate and pregnancy were developed. All the parameters in the models were dichotomized and intra-and interindividual variability of the parameters were assessed. Although the sperm parameters showed good correlations with IVF when correlated separately, the only essential parameter in the multivariate model was morphology. The enormous intra-and interindividual variability of the values was striking. In conclusion, our data indicate that the andrological status at the end of the respective treatment does not necessarily represent the status at the time of IVF. Despite a relatively low correlation coefficient in the logistic regression model, it appears that among the parameters tested, the most reliable parameter to predict fertilization is normal sperm morphology. (Reprod Med Biol 2005; 4: 7-30) Key words: assisted reproduction, high intra-and interindividual variability, multivariate approach, prediction of outcome of IVF, sperm functions.

No abstract

The significant increases of PMN-elastase levels as marker of male genital tract inflammation in older men appear to be indicative of age-related changes in local immunoregulatory mechanisms. Because there is no association of PMN elastase with sperm motility, a direct inhibitory effect of this enzyme can be excluded.

The presence of excess leucocytes in the semen has been associated with male infertility. According to the WHO, concentrations of more than 106 leucocytes ml-1 are considered as leucocytospermia, indicating genital tract infections. Up to now, no consensus has been achieved on how leucocytes should be quantified in semen. Using the peroxidase staining and monoclonal antibodies to CD15, CD45 and CD68, we found significant differences between the detection methods. Only 47.4% of the semen samples that were assessed as leucocytospermic by CD45 were identified as such by peroxidase staining. The concentration of peroxidase-positive cells was significantly correlated with polymorphonuclear granulocyte (PMN) elastase (P < 0.0001). However, a negative correlation of peroxidase-positive cells with the sperm concentration was only found in oligozoospermic patients (P < 0.0001). Moreover, the slightly positive correlation with normal sperm morphology seems to be applicable only in cases of oligozoospermia. Significant negative correlation of the number of peroxidase-positive cells were found for both maximal inducible acrosome reaction (P = 0.0219) and the inducibility of acrosome reaction (P = 0.0370), indicating a rather deleterious effect of leucocytes on this important sperm function. Concerning the result in the in vitro fertilization programme, none of the examined parameters (PMN elastase, concentration of round cells and peroxidase-positive cells) showed a correlation with either fertilization or pregnancy. This result seems to be reasonable as severely damaged spermatozoa and leucocytes are eliminated from the ejaculate by different sperm separation methods. Interestingly, a significant negative correlation of the TUNEL assay as a measure of sperm DNA fragmentation was found only with pregnancy (P = 0.006) but not with fertilization. As DNA fragmentation can also be caused by ROS that are generated by leucocytes, this causality should not be neglected.