Sperm membrane integrity in fresh and frozen-thawed canine semen samples: a comparison of vital stains with the NucleoCounter SP-100

The objective of this study was to assess sperm membrane integrity in canine semen samples using three different vital stains and the NucleoCounter SP-100 (NC). In addition, the occurrence of half-stained sperm heads, the influence of investigator, and storage-related artifacts using stained smears were examined. Forty fresh (30 dogs) and 40 frozen-thawed (28 dogs) canine semen samples were analyzed. The vital stains eosin (E), eosin-nigrosin (EN), and bromphenolblue-nigrosin (BN) were compared. Two smears per stain were prepared and a total of 200 sperm per slide were classified using bright field microscopy. Each slide was examined twice by three investigators. Spermatozoa with completely red (E, EN) or blue (BN) stained sperm heads were classified as "dead". Half-stained sperm heads were counted separately. Sperm concentration and viability were determined using the NC. The NC works with a built-in fluorescence microscope using propidium iodide as a fluorescence dye. Statistical analysis for comparison of results was made using mean values with standard deviation, Bland-Altman plot and coefficient of variation (CV). Staining with E led to a significant higher percentage of dead sperm compared with EN and BN (P < 0.05), which gave comparable results. Vital stains revealed higher CVs (range 8.8%-32.1%) than the NC (<6.5%). Interobserver viability ranged from 17.5% to 45.4% and was within the same range between stains. If only completely stained sperm heads were considered, best agreement was found between the E and the NC. In case of EN and BN, inclusion of half-stained sperm heads reduced the difference compared with NC. In general, the agreement between methods was better in samples with a low percentage of dead spermatozoa. In smears of fresh semen stored up to 3 months, no increase in the percentage of dead spermatozoa could be observed. In some smears of frozen-thawed samples stained with E (n = 12) or BN (n = 2), all previously unstained spermatozoa were additionally stained during storage. In conclusion, all vital stains examined in this study technically can be used for differentiation between live and dead spermatozoa in canine semen samples, but the relatively high CVs have to be kept in mind. It is recommended to examine smears of frozen-thawed semen soon after preparation. Half-stained sperm heads should be counted as live sperm in E-stained smears. The NC allows assessment of sperm concentration and viability with a reasonable repeatability.