Determinantion of testosterone concentration in semen of men with normal or subnormal sperm counts and after vasectomy.

Semen from 58 male subjects, aged 22 to 50, was assayed on an individual basis to determine whether T was present in it. Of the subjects examined 23 were normospermic, 14 oligospermic and 9 azoospermic; 12 men had undergone vasectomy were also included in the study. In 39 of the subjects plasma testosterone was estimated. A competitive protein binding technique was employed for T assays while dried extracts of semen were examined by combined gas-chromatography-mass spectrometry and mass fragmentography. Measurable amounts of T were detected in all seminal specimens assayed. This was confirmed by gas chromatography-mass spectrometry which showed a spectrum suggestive of T. The ratio of unconjugated to conjugated steroid in semen was found to be approximately 1:10. Levels of unconjgated T were similar to those found in plasma of normally menstruating women. The mean seminal concentration of unconjugated T (+/-SD) in the specimens assayed was 0.71 ng/ml+/-0.08 for the normospermic, 0.79+/-0.14 for the ezoospermic, 0.69+/-0.09 for the oligospermic, but only 0.38+/-0.04 for the vasectomized subjects. Plasma levels for this androgen were within the range found in normal men of comparable age. Significant correlation between plasma and seminal T concentration could not be demonstrated and there was no correlation between either of the above parameters and the seminal volume, the number, abnormal form percentage and the motility of spermatozooa in the normo--or oligospermic group. However, when the two groups were pooled into one, significant correlations were found between plasma, (but not seminal T concentration) and the seminal characters examined, perhaps suggesting the number of specimens from the groups should be increased to obtain valid data. Administration of human chorionic gonadotropin produced a marked plasma response as well as a rise of seminal T levels in 3 normospermic subjects whereas cyproterone acetate caused reduction of plasm T levels but had no consistent effect on the seminal concentration of ts steroid although the sensitivity of the seminal method may not have detected smaller changes at this level.