Abstract: Objective To identify influencing factors of prognosis of extremely low birth weight infants (ELBWI). Methods Clinical data of ELBWI admitted during the study period were retrospectively collected, and their in-hospital outcomes were recorded. Infants who received active treatment were enrolled in active treatment group, while those who were given up treatment due to family reasons in the absence of severe complications were enrolled in treatment-withdrawal group. According to gestational age, the active treatment group was further subdivided into four subgroups: 23-25⁺⁶ weeks, 26-27⁺⁶ weeks, 28-29⁺⁶ weeks, and ≥30 weeks. Based on year of birth, infants were divided into 2010–2019 group and 2020–2024 group. The Aalen-Johansen competing risk model was used to estimate the cumulative incidence of severe complications. Multivariate logistic regression analysis was performed to identify independent risk factors for mortality. Cox proportional hazards regression model was used to analyze factors associated with survival without major complications. Results A total of 151 ELBWI were enrolled, including 79 males (52.3%) and 72 females (47.7%). The median gestational age was 27.4 (26.4, 29.0) weeks, and the median birth weight was 840 (750, 930) g. There were 14 cases (9.3%, 14/151) in treatment-withdrawal group, 137 cases (90.7%, 137/151) in the active treatment group. In the active treatment group, 16 infants (11.7%, 16/137) died or were withdrawn from treatment due to severe complications, and 121 infants (88.3%, 121/137) survived at discharge. The survival rates of rescue by gestational age subgroups were as follows: 75.0% (21/28) in the 23-25 weeks group, 90.9% (40/44) in the 26-27 weeks group, 93.6% (44/47) in the 28-29 weeks group, and 88.9% (16/18) in the ≥30 weeks group. Among the 137 actively treated infants, the incidences of moderate-to-severe bronchopulmonary dysplasia, sepsis, severeretinopathy of prematurity, severe intraventricular hemorrhage（IVH）, necrotizing enterocolitis (≥ stage II), and periventricular leukomalacia were 60.6% (83/137), 30.7% (42/137), 12.4% (17/137), 13.1% (18/137), 8.8% (12/137), and 1.5% (2/137), respectively. Compared with the 2010–2019 group, the 2020–2024 group had a higher rate of survival without major complications, a shorter time to full enteral feeding, and a lower incidence of moderate-to-severe BPD (P=0.049, P=0.001, P=0.041, respectively). Multivariate logistic regression analysis revealed that higher gestational age was a protective factor against mortality (OR=0.73, 95% CI: 0.57–0.93, P=0.013), while severe IVH was an independent risk factor for mortality (OR=6.36, 95% CI: 1.45–27.92, P=0.015). Cox regression analysis showed that gestational age was an independent protective factor for survival without major complications (HR=1.24, 95% CI: 1.03–1.51, P=0.026), indicating that each additional week of gestational age was associated with a 24% increase in the likelihood of survival without major complications. Sex (HR=1.35, P=0.558) and adequate antenatal dexamethasone use (HR=0.83, P=0.654) showed no significant association with survival without major complications. Conclusion Gestational age is an independent predictor of prognosis in ELBWI. A comprehensive management strategy oriented toward "improving quality of survival" should be implemented to maximize the quality of survival while ensuring the survival rate of ELBWI.