SLIDE PREPARATION AND STAINING PROCEDURES FOR RELIABLE RESULTS USING COMPUTERIZED MORPHOLOGY

The purpose of this study was to standardize slide preparation and sta ining procedures to improve the efficiency and effectivity of the IVOS system on normal sperm morphology readings with regard to the strict criteria. Semen samples from patients attending the Reproductive Biolo gy Unit, Tygerberg Hospital, were used. In experiment 1, five differen t Diff-Quik staining procedures, including the standard procedure, wer e evaluated on each of 22 patients and the effect of slide preparation within 1 h or more than 5 h after collection and the effect of immedi ate fixation versus fixation after 24 h were observed. In experiment 2 , the manual evaluation time per slide (n = 20) by two technicians was compared with the time taken by computer. In experiment 1 the median % normal for the 5 different staining procedures was 6, 6.5, 9.5, 8.5, and 5.5%. No significant difference was found between the different s taining procedures (p = .60, nonparametric Friedman test). In experime nt 2 the mean time for manual assessment by two technicians was 3 min: 6 s and 3 min:53 s per slide as compared to 4 min:39 s by computer. Fo r experiment 1, slides can be prepared immediately or after 5 h. Fixat ion time also does not interfere with the computer's ability to identi fy normal forms. For experiment 2, the IVOS system is competitive rega rding assessment time. Standardization of optimum staining procedures is important to ensure repeatability and comparability. Therefore, sli des should be prepared immediately after liquefaction and fixed immedi ately after air drying.