Some studies show that children born after ICSI with non-ejaculated sperm are at increased risk of birth defects, other studies hold the opposite view. Does neonatal outcome including congenital malformations in children born after ICSI with percutaneous epididymal sperm aspiration (PESA) and testicular sperm aspiration (TESA) differ from neonatal outcome in children born after ICSI with ejaculated sperm? In this study, we examined the data from our IVF center from 2006 to 2016, to compare neonatal outcomes and rates of congenital malformations in children born after ICSI with different sperm origin. The results showed the clinical pregnancy rate and implantation rate of non-ejaculated sperm group were significantly higher (P < 0.001) than ejaculated sperm group. There were 775 clinical pregnancies from non-ejaculated sperm group and 2,486 clinical pregnancies from ejaculated sperm group. Most of the clinical pregnancy outcomes were comparable between non-ejaculated sperm group and ejaculated sperm group (p > 0.05): the miscarriage rate per transfer, ectopic pregnancy rate per clinical pregnancy, induced abortion rate per clinical pregnancy and fetal deaths per clinical pregnancy. However, the live delivery rate per transfer of non-ejaculated sperm group was significantly higher than that of ejaculated sperm group (45.4% vs 36.7%, P < 0.001). Moreover, the comparison between the epididymal sperm, testicular sperm and ejaculated sperm groups showed there were no difference in the incidence of congenital malformations of babies live birth. Among singleton gestation live births, there were more girls than boys in both non-ejaculated sperm and ejaculated sperm group. In conclusion, the present study clearly showed no statistical increased risk in neonatal outcomes of newborns were found in the ICSI treatment with epididymal or testicular sperm. It may provide information for consultation for ICSI treatment in PESA or TESA patients.