Reliability of estimates of diploid human spermatozoa using multicolour fluorescence in-situ hybridization [published erratum appears in Hum Reprod 1997 May;12(5):1118]

Multicolour fluorescence in-situ hybridization (FISH) analysis permits distinction between disomic and diploid spermatozoa. Thus estimates of the frequency of diploid spermatozoa can be obtained for human semen samples. The issue of the accuracy and reliability of these diploidy estimates has been addressed by analysing diploidy frequencies in 10 men using the same sperm sample to estimate diploidy twice-once during two-colour FISH analysis of disomy for chromosomes 1 and 12 and a second independent analysis of three-colour FISH for disomy estimates for chromosomes X and Y (with chromosome 1 used as the autosomal control). A minimum of 10,000 spermatozoa per hybridization per male was counted for a total of over 200,000 spermatozoa analysed. The mean frequency of diploid spermatozoa was 0.13% for the autosomal study and 0.14% for the sex chromosomal study, which were not significantly different. One donor had extremely divergent values of diploidy in the two studies. Analysis of values in the other nine donors demonstrated no significant difference in the two diploidy estimates. These results indicate that the FISH technique is an accurate and reliable method for determining diploid frequencies in human spermatozoa.