RESUMEN
PURPOSE: To identify the sperm preparation procedure that selects the best sperm population for medically assisted reproduction. METHODS: Prospective observational study comparing the effect of four different sperm selection procedures on various semen parameters. Unused raw semen after routine diagnostic analysis was split in four fractions and processed by four different methods: (1) density gradient centrifugation (DGC), (2) sperm wash (SW), (3) DGC followed by magnetic activated cell sorting (MACS), and (4) using a sperm separation device (SSD). Each fraction was analyzed for progressive motility, morphology, acrosome index (AI), and DNA fragmentation index (DFI). RESULTS: With DGC as standard of care in intraclass correlation coefficient analysis, only SSD was in strong disagreement regarding progressive motility and DFI [0.26, 95%CI (- 0.2, 0.58), and 0.17, 95%CI (- 0.19, 0.45), respectively]. When controlling for abstinence duration, DFI was significantly lower after both MACS and SSD compared to DGC [- 0.27%, 95%CI (- 0.47, - 0.06), p = 0.01, and - 0.6%, 95%CI (- 0.80, - 0.41), p < 0.001, respectively]. Further comparisons between SSD and MACS indicate significantly less apoptotic cells [Median (IQR) 4 (5), 95%CI (4.1, - 6.8) vs Median (IQR) 5 (8), 95%CI (4.9, - 9.2), p < 0.001, respectively] and dead cells [Median (IQR) 9.5 (23.3), 95%CI (13.2, - 22.4) vs Median (IQR) 22 (28), 95%CI (23.1, - 36.8), p < 0.001, respectively] in the SSD group. CONCLUSION: The selection of a population of highly motile spermatozoa with less damaged DNA from unprocessed semen is ideally performed with SSD. Question remains whether this method improves the embryological outcomes in the IVF laboratory.