目的 探讨小儿危重病例评分（PCIS）、儿童死亡风险评分（PRISM Ⅲ）、小儿序贯器官衰竭评估评分（pSOFA）、肺水肿放射学评分（RALE）、急性肺损伤评分（LIS）对儿童急性呼吸窘迫综合征（pARDS）预后评估的应用价值。方法 回顾性分析2015年1月至2021年12月入住儿科重症监护病房（PICU）诊断为ARDS患儿的临床资料。结果 纳入80例ARDS患儿，均存在脓毒症和多器官功能障碍，男45例、女35例，中位年龄12.5（4.0~36.3）月龄。存活组50例、死亡组30例（住院期间死亡16例，放弃后死亡14例）。与存活组相比，死亡组氧合指数（OI）较高，转入PICU及诊断ARDS时pSOFA、PRISM Ⅲ评分较高，诊断ARDS时RALE、LIS评分较高，应用血管活性物质及输血比例较高；总住院时间、PICU住院时间、机械通气时间较短，肺内因素比例、PO₂/FiO₂、PCIS评分较低，差异均有统计学意义（P<0.05）。与不合并基础疾病组比较，合并基础疾病组诊断ARDS时的pSOFA、PRISM Ⅲ、RALE、LIS评分较高；PCIS评分较低，差异均有统计学意义（P<0.05）。诊断ARDS时PCIS、pSOFA、PRISM Ⅲ、RALE、LIS评分预测ARDS患儿死亡的AUC分别为0.73、0.89、0.83、0.80、0.82。Hosmer-Lemeshow 拟合优度检验显示PCIS预测死亡率与实际死亡率的拟合效果最优（χ²=4.16，P=0.656）。结论 PCIS、pSOFA、PRISM Ⅲ、RALE、LIS评分均对ARDS患儿预后具有良好的预测能力，pSOFA评分预测价值最大，PCIS评分的预测病死率与实际病死率一致性最佳。

Objective To explore the application value of pediatric critical illness score (PCIS), pediatric risk of mortality (PRISM) Ⅲ score, pediatric sequential organ failure assessment (pSOFA) score, radiographic assessment of lung edema (RALE) score and acute lung injury score (LIS) in the prognosis evaluation of pediatric acute respiratory distress syndrome (pARDS). Methods The clinical data of children diagnosed with ARDS admitted to the pediatric intensive care unit (PICU) from January 2015 to December 2021 were retrospectively analyzed. Results A total of 80 ARDS children with sepsis and multiple organ dysfunction were enrolled, including 45 boys and 35 girls, and the median age was 12.5 (4.0-36.3) months. There were 50 patients in the survival group and 30 patients in the death group (16 died during hospitalization and 14 died after abandonment). Compared with the survival group, the death group had higher oxygenation index (OI), higher pSOFA and PRISM Ⅲ scores at transferring to PICU and diagnosing ARDS, higher RALE and LIS scores at diagnosing ARDS, and higher proportion of vasoactive substances application and blood transfusion. Furthermore, compared with the survival group, the death group had shorter total hospital stay, PICU stay and mechanical ventilation time, and lower proportion of intrapulmonary factors, PO₂/FiO₂ and PCIS scores. The differences were statistically significant (P<0.05). Compared with the group without underlying diseases, the scores of pSOFA, PRISM Ⅲ, RALE, and LIS were higher and the scores of PCIS were lower at the diagnosis of ARDS in the group with underlying diseases, the differences were statistically significant (P<0.05). The AUCs of PCIS, pSOFA, PRISM Ⅲ, RALE, and LIS at the diagnosis of ARDS in predicting the death of ARDS children were 0.73, 0.89, 0.83, 0.80 and 0.82, respectively. Hosmer-Lemeshow goodness of fit test showed that PCIS had the best fitting effect between predicted mortality and actual mortality (χ²=4.16, P=0.656). Conclusions PCIS, pSOFA, PRISM Ⅲ, RALE and LIS scores all have good predictive ability for the prognosis of children with ARDS, and pSOFA score had the highest predictive value. PCIS had the best fitting effect between predicted mortality and actual mortality.