Effects of different staining and washing procedures on the results of human sperm morphology evaluation by manual and computerised methods.

The purpose of this study was to investigate the effect of different staining and washing procedures on the results of human sperm morphology evaluation by manual and computerised methods. Furthermore, it was intended to find the staining and washing combination which would provide optimal readability for computer-assisted sperm morphology evaluations. In phase one, four staining methods were evaluated for smears prepared from the resulting samples following a two times washing procedure. In phase two, 20 semen samples were used to compare the Diff-Quik and Papanicolaou staining methods, following one and two washes. All manual readings, of Papanicolaou and Diff-Quik stained smears, were comparable with each other, with means between 7.3% and 7.9% normal spermatozoa. All the manual readings were also comparable to the computer readings of the Diff-Quik slides following one and two washes with means of 9.0% and 5.9%, respectively. However, due to the higher computer readings found for the Papanicolaou stained smears, with means of 13.9% and 13.5% following one and two washes, respectively, a statistically significantly difference between overall computer and manual readings was found (Wilks' Lamda, P = 0.0002). Taking all data into consideration, it could be concluded that the one wash Diff-Quik stained smears was the optimal preparation method for computerised sperm morphology evaluation, comparing favourably with manual evaluations.