Effect of e-learning training on technical performance for bovine sperm morphology analysis in a multi-laboratory...

Although sperm morphology analysis is a fundamental component of bovine semen evaluation, substantial variations in results within and across laboratories have been reported. The objective of the present study was to evaluate the effect of e-learning training to standardize morphology analysis in a multi-laboratory setting. In Experiment I, the performance of 40 technicians from 10 laboratories was evaluated using digital sperm images. Assessment score improved (P < 0.05) shortly after training (85.1 ± 1.3 %; mean ± SEM) when compared to pre-training baseline (78.3 ± 1.8 %) but was not significantly different from baseline eight months after training (83.8 ± 1.7 %). In Experiment II, the performance of eight technicians from six laboratories was evaluated using reference frozen-thawed semen samples. Overall intra-technician CV for normal sperm, and primary and secondary defects were 4.0 ± 4.3 %, 22.9 ± 25.2 %, and 16.8 ± 16.0 %, whereas intra-batch CV were 6.2 ± 4.4 %, 33.7 ± 19.8 %, and 24.0 ± 12.4 %, respectively. The intra-technician CV for normal sperm decreased shortly after training when compared to pre-training baseline and remained low six months after training. The intra-batch CV for normal sperm decreased shortly after training but was not significantly different from baseline six months after training. When Bland-Altman plots were evaluated, the mean difference of individual results from the overall mean was close to zero (-2.1 to 1 %) for all morphology classification categories during all test periods. In conclusion, e-learning training improved technician performance for bull sperm morphology analysis. Precise and accurate results for normal sperm morphology can be obtained with the use of standardized procedures when performed by trained and experienced technicians, but biannual re-training and assessment might be required to maintain optimal technical performance. In addition, simple sperm morphology classification systems (normal vs. abnormal) may be more robust than complex defect categorizations for quality control in semen processing centers.