By D. Zaritskiy and W.L. Flowers, North Carolina State University Department of Animal Science
Use of computer-assisted semen analysis systems is becoming increasingly commonplace in boar studs. Some of the main reasons for this are that they are fast, quantitative and accurate, and can provide information in addition to just motility. This additional information is useful for trouble-shooting potential problems with boars or with semen collection and extension.

Nevertheless, as with most technologies, CASA measurements are subject to potential external biases. These can be associated with the manner in which technicians prepare semen for evaluation or how they conduct the evaluation itself. The degree to which several of these factors can bias CASA was studied in a series of experiments. Each of these experiments was conducted with 14 boars collected weekly over an eight-week period and used the split-ejaculate technique so that each boar and ejaculate served as its own control.

The first external factor that was studied was the effect of the time interval between loading the slide and when the analysis was performed. For this experiment, slides were loaded and measurements were recorded immediately after the evaluation chamber was completely filled with the sample (time 0) and then again 30, 60, 90 and 120 seconds later. Mean motility; progressive motility; and curvilinear velocity for all ejaculates at time 0 during the study were 91.5%, 78.0% and 122.2 um/s, respectively. These and all other estimates of motility decreased significantly during the first 60 seconds after loading the slide. Motility was reduced by 7% while progressive motility and curvilinear were 15% and 17% lower, respectively.

The second external factor that was studied was dilution rate. For this experiment, CASA was conducted on the neat ejaculate and samples of the neat ejaculate diluted 1:1, 1:2 and 1:3 with BTS semen extender. Based on the results of the first experiment all readings were taken within 30 seconds after evaluation chambers were filled. The effect of dilution was variable across all the CASA variables studied. However, the highest readings arithmetically were consistently observed with a 1:1 dilution. Motility was similar statistically between the neat sample and the 1:1 dilution, but decreased linearly with the higher dilutions. In contrast, a significant increase occurred when the neat sample was diluted 1:1 for progressive motility, curvilinear velocity and most other CASA variables, but then decreased as the dilatation rate increased. Dilution rate had a much lower impact on CASA variables compared with time. Differences between the highest and lowest readings among the undiluted and diluted samples were only 4 to 7%.

The third external factor that was studied was the technician who prepared the slide and performed the CASA. For this experiment, five different individuals with different levels of experience prepared and performed CASA on each of the ejaculates. The most experienced had over 1,000 hours of using CASA while the least experienced had never loaded slides or performed CASA prior to the study. All ejaculates were diluted 1:1 prior to being loaded onto the evaluation slides and all technicians were instructed to analyze their slide within 30 seconds of when the chamber was completely filled with the sample. There was no effect of technician on any of the CASA measurements. In fact, arithmetically, there was less than a 2% difference between the highest and lowest estimates for motility and progressive motility among the five different technicians who participated in the study.

In summary, it appears that the time interval between when the chamber on the slide was completely filled and when readings were taken had the most significant effect on CASA measurements. For all variables studied, this relationship was an inverse one: as time increased all the various estimates of motility decreased. While a difference of 7 to 10% may seem small for a boar that produces an ejaculate with 100 billion total sperm then would equate to a reduction of about 7 billion to 10 billion sperm or two to three insemination doses from each collection. Dilution rate had variable effects on CASA measurements which were much less pronounced than those observed for time.

However, a 1:1 dilution consistently produced the highest estimates. Finally, there was no effect of technician even when significant differences in experience with CASA was present. It is important to note that there was some standardization among technicians since all ejaculates used were diluted 1:1 and technicians were instructed to obtain the measurements within 30 seconds of analysis chambers being fully loaded.