Standardization and comparability of CASA instruments.

Thirty human semen specimens were analyzed using a standard manual method, then videotaped and reanalyzed using two different computer-aided sperm analysis (CASA) instruments (the HTM system, Hamilton-Thorn Research, Danvers, MA, and the CTS system, Motion Analysis Corp., Santa Rosa, CA). Videotaped specimens were analyzed by CASA for 5 frames for sperm concentration (CON) and percent motility (MOT), and for 15 frames for kinematic variables (straight-line velocity, VSL; curvilinear velocity, VCL; linearity, LIN; and amplitude of lateral head displacement, ALH). Machine parameter settings for the two instruments were matched as closely as possible. CASA values were compared with each other for all measures and with manual results for sperm count and the percentage of motile sperm. Results show: 1) HTM and CTS average values for CON are not different from manual measures for the 5- or 15-frame analysis, but slight differences are seen between CTS and HTM; 2) average values for MOT for the 5-frame analysis are higher than the 15-frame analysis for both instruments, but the average manual measurement for percent motility is much higher than any CASA value; 3) average VSL and LIN are slightly higher for HTM than CTS, but pair-wise comparison shows a high degree of concordance between the instruments; and 4) the mean values for VCL and ALH are equal for the two instruments, and there is a close concordance for the pair-wise comparison for VCL; however, pair-wise comparison of ALH reveals significant differences between the instruments. Overall, the differences seen between these instruments are slight, and are probably not biologically or clinically significant.