Objective: The main objective is to review the overall result and impact of preoperative testosterone level on sperm retrieval rate (SRR) by microdissection testicular sperm extraction (micro-TESE) in patients with nonobstructive azoospermia (NOA). Materials and Methods: We retrospectively reviewed the files of patients who underwent micro-TESE for NOA from August 2013 to December 2014. All patients were evaluated with history, physical examination, and hormonal assessment. Patients who had previous micro-TESE, obstructive azoospermia, or who took hormone therapy were excluded from the study. Patients were classified into two groups. Group A included patients who had low testosterone (<10 nmol/L), and Group B included patients with normal testosterone (>10 nmol/L). The primary endpoint was to review the overall results of the procedure and the impact of preoperative testosterone level on sperm retrieval. Results: A total of 264 patients with NOA underwent micro-TESE. Group A included 133 patients with low testosterone (<10 nmol/l) with a median age of 36 ± 6.59 years, and Group B included 131 patients with normal testosterone (>10 nmol/L) with a median age of 33 ± 7.88 years ( P = 0.1350). There was no significant difference in follicle-stimulating hormone ( P = 0.2467), luteinizing hormone ( P = 0.1078), prolactin ( P = 0.5619), and testicular volume ( P = 0.4052), whereas a significant difference was found in testosterone level ( P = 0.0001) in both groups. Overall, sperm were successfully retrieved in 48.8% of men. SRR in Group B was significantly higher (57.25%) than that in Group A (40.60%) ( P = 0.0068). SRR in patients with Sertoli-cell-only pathology was 30.35%, hypospermatogenesis was 89.74%, and maturation arrest was 32.43%. Conclusion: Micro-TESE is a successful and safe procedure in NOA patients with a poor prognosis. Preoperative testosterone level has a significant impact in the SRR by micro-TESE.