The aim of this study was to relate retrospectively in 215 percutaneous large needle aspiration biopsy (LNAB) of the testicles the number of seminiferous tubules (ST) per testicular biopsy, from differently sized testicles, to the size of the needle used in order to obtain information useful for selecting the needle size according to the testicular size. Clinical, ultrasonographic and pathologic data of 134 patients with nonobstructive azoospermia examined with LNAB were retrospectively analyzed. One hundred and sixty-three testicles with volume≥10 ml, 39<10 ml and≥6 ml, 13<6 ml underwent 215 LNABs. In the larger testicle group the prevalence of adequate biopsies (ST≥20) varied from 83% to 50% with a statistically significant difference among the needle groups (P=0.028) and being 83% in the biopsies with the 18 gauge (g.) needle. The mean number of ST obtained with the 18 g. needle was significantly higher than that obtained with the other needles (P=0.000). In the medium volume testicle group the adequate biopsy incidence varied from 20% to 83.3% with statistical significance among the various needle groups (P=0.042) being 83.3% in the biopsies with the 22 g. needle. In the smaller testicle group the mean number of ST obtained with the 22 g. needle was significantly higher than with the 20 g. needle (71 versus 25, P=0.011). In the patients with nonobstructive azoospermia, the 18 g. needle seemed to offer the best performance in the larger testicles, while in the medium and lower sized testicles the smaller 22 g. needle gave results comparable to or slightly better than the other larger needles.