To investigate the outcome of intracytoplasmic sperm injection with fresh and cryopreserved-thawed testicular spermatozoa in the first cycle in patients with obstructive azoospermia (OA) and non-obstructive azoospermia (NOA), a total of 90 cases, 48 OA and 42 NOA were studied. All patients underwent sperm retrieval by testicular sperm extraction (TESE) while their wives received conventional ovarian hyperstimulation. The hormone levels, testicular histology, the rates of sperm retrieval, fertilization, implantation and pregnancy were analysed and evaluated. This study and other four similar studies were subjected to meta-analysis. Sperm retrieval was successful in 100% OA and 61% NOA. Fresh spermatozoa were used in 87.5% and 92.4% of OA and NOA cases respectively; while cryopreserved-thawed spermatozoa were used in 12.5% and 7.6% of OA and NOA, respectively. The fertilization, implantation and clinical pregnancy rates were 65.5%, 15% and 25% respectively in OA group, and 54.2%, 5% and 23.1% respectively in NOA group. Sperm status (fresh or thawed), male partner's age, female age and male serum follicle-stimulating hormone had no significant effect upon fertilization rate, implantation rate, or pregnancy rate per embryo transfer. The results of meta-analysis indicate that there is no statistically significant difference in clinical pregnancy rates between the two groups. There was a significantly higher fertilization rate among OA patients in all analysed studies (95% CI = 14.29-15.71, d.f. 832, T = 1.96). In conclusion, although the fertilization rate was significantly higher in the OA group in our study and from the given meta-analysis, there were some differences as regards pregnancy rates. Although the overall effect was more or less similar pregnancy rates in both subtypes of azoospermia, this may not be true if non-male infertility variables were controlled for in all studies.