Objective To explore the risk factors of premature infants death. Methods The medical records of hospitalized premature infants admitted to West China Second University Hospital of Sichuan University between January 2015 and December 2022 were collected. Premature infants were divided into the death group and the non-death group (control group) based on discharge diagnosis of death. Parturient and premature infants related information were collected, and the disease classification and diagnosis of premature infants were analyzed. Results A total of 13 739 premature infants were included, with 53 deaths and a mortality rate of 3.85‰ (53/13 739). The ages of death were 1-49 days, and the median age of death was (9.68±9.35) days. According to the matching method, 212 premature infants were ultimately included. Among them, there were 53 premature infants in the death group and 159 premature infants in the control group. Compared with the control group, premature infants in the death group had lower gestational age, birth weight, lower 1-minute Apgar scores, lower 5-minute Apgar scores and shorter hospital stay (P<0.05), and received more delivery interventions (P<0.05). There was no statistically significant difference in other indicators between the two groups of premature infants (P>0.05). A total of 212 parturient were included. Among them, there were 53 parturients in the death group and 159 parturients in the control group. The use rate of prenatal corticosteroids in the control group was higher than that in the death group (55.35% vs. 54.72%). There was no statistically significant difference in other related factors between the two groups of parturient (P>0.05). The results of logistic regression analysis showed that longer hospital stay [odds ratio (OR)=0.891, 95% confidence interval (CI) (0.842, 0.943), P<0.001], prenatal use of corticosteroids [OR=0.255, 95%CI (0.104, 0.628), P=0.003] reduced the risk of premature infant death. However, tracheal intubation [OR=10.738, 95%CI (2.893, 39.833), P<0.001] increased the risk of premature infant death. Conclusions Clinicians should pay attention to prenatal examination of newborns and pay attention to evaluation of newborn status. Obstetricians and neonatologists should make joint plans for women with high risk factors for preterm delivery. During the hospitalization, after the diagnosis is clear, standardized treatment should be carried out in strict accordance with the guidelines for systemic diseases and expert consensus.