Testicular biopsy and intracytoplasmic sperm injection (TESE/ICSI) became standard treatment for patients with non-obstructive azoospermia. However, TESE may not always be successful, furthermore, it may has significant complications. Thus , predicting TESE results is of utmost importance to safeguard the patient against invasive diagnostic procedures with known potential complications. The aim of this study is to find the value of various clinical and laboratory parameters, particularly identification of seminal spermatids using May-Grünwald-Giemsa (MGG) stain and phase contrast optics in predicting TESE result. One hundred and one patients with non-obstructive azoospermia (NOA) underwent diagnostic TESE through multiple testicular sampling after clinical evaluation, serum FSH measurement and identification of seminal spermatids and spermatocytes using MGG stain and phase-contrast optics. Result of TESE were compared with previous parameter in addition to histopathology. Sperms were retrieved in 50.5% of patients. Primary spermatocytes were detected by MGG in 68.6% of TESE +VE cases and in 38% of TESE –VE cases, while round spermatid were identified in 58.9% of TESE +VE cases and in 2% of TESE-VE cases, The MGG specificity of spermatid detection in predicting TESE outcome was 98%. Using phase-contrast optics, round spermatids were identified in 43.1% of TESE +VE cases and in 4% of TESE-VE cases, the specificity of phase-contrast spermatid identification was 96% in predicting TESE outcome. Testicular histopathology was in general an inaccurate parameter. Testicular volume , serum FSH level and age had poor predictive value. We conclude that detection of round spermatids in semen by MGG stain is a simple, cost effective, commercially available and highly predictive for TESE result. Moreover using phase-contrast optics in identification of round spermatid in semen is a good parameter for TESE prediction but less accurate than MGG stain.