Purpose: In men considered to have azoospermia by routine semen analyses sperm may be identified after centrifuging the semen. Because these sperm may be used for intracytoplasmic sperm injection, we describe our technique and findings of sperm pelleting.
Materials and methods: Semen centrifugation for sperm pellet analysis was performed in 140 consecutive men in whom no sperm was identified on routine semen analysis and who were categorized as having obstructive or nonobstructive azoospermia. Obstructive azoospermia was defined as failed vasectomy reversal, failed reconstruction for congenital vasal or epididymal occlusion, or an acquired obstruction unrelated to ejaculatory duct obstruction. Patients with congenital absence of the vas deferens or who had undergone vasectomy were not included in the study. Nonobstructive azoospermia was defined as moderate to severe testicular atrophy with markedly elevated serum follicle-stimulating hormone (greater than 3 times normal), or a testicular biopsy that revealed maturational arrest, severe hypospermatogenesis or the Sertoli-cell-only pattern. Obstructive and nonobstructive azoospermia were present in 70 men who provided 109 samples and 70 who provided 103, respectively.
Results: Motile and nonmotile sperm was identified in 13 of the 70 patients (18.6%) with obstructive and in 16 of the 70 (22.8%) with nonobstructive azoospermia. Pellet variability, that is the absence of sperm in 1 specimen and its presence in another from the same patient, was noted in 7 of the 17 men (41.2%) with obstructive and 2 of the 17 (11.8%) with nonobstructive azoospermia (not statistically significant). Motile sperm was present in the pellets of 6 of the 70 men (8.6%) with obstructive and 15 of the 70 (21.4%) with nonobstructive azoospermia. The median number of motile sperm was lower in the obstructive than in the nonobstructive group (0 sperm in 17 samples versus 5 sperm in 41 samples, p <0.001). The median value of 0 in the obstructive azoospermia group reflects the finding that 9 of the 17 samples did not contain motile sperm. Similarly the median number of nonmotile sperm was lower in the obstructive than in the nonobstructive group (5 versus 8 sperm).
Conclusions: We demonstrated the presence of motile and nonmotile sperm in a significant number of men considered to have azoospermia by routine semen analysis. Semen centrifugation (sperm pelleting) should be performed in all men considered to have this condition by routine semen analysis, especially those with testicular failure and those in whom intracytoplasmic sperm injection is possible.