目的 评估儿童序贯器官衰竭评分（Pediatric Sequential Organ Failure Assessment, pSOFA）、第三代小儿死亡指数评分（Pediatric Index of Mortality 3, PIM⁃3）、儿童第四代死亡风险评分（Pediatric Risk of Mortality ScoreⅣ, PRISM Ⅳ）、菲尼克斯脓毒症评分（Phoenix Sepsis Score, PSS）、菲尼克斯脓毒症评分⁃8（Phoenix Sepsis Score⁃8, Phoenix⁃8）、中性粒细胞计数与淋巴细胞和血小板计数比值（ratio of neutrophil count to lymphocyte and platelet count, N/LPR）及构建的列线图预测模型对脓毒症患儿28 d死亡的预测价值。 方法 回顾性分析2017年1月—2025年1月郑州大学第三附属医院儿科重症监护室收治的267例脓毒症患儿，分为生存组（208例）和死亡组（59例）。比较两组临床指标差异，通过单因素及多因素logistic回归筛选28 d死亡的独立危险因素并构建列线图模型，采用受试者操作特征曲线、校准曲线及决策曲线评估列线图模型的预测效能。 结果 死亡组pSOFA、PIM⁃3、PRISM Ⅳ、PSS、Phoenix⁃8评分及N/LPR、乳酸水平显著高于生存组（P<0.05）。多因素logistic回归分析确定pSOFA、PRISM Ⅳ、PSS评分及N/LPR为脓毒症患儿28 d死亡的独立危险因素。基于这些危险因素构建的预测28 d死亡风险的列线图模型的受试者操作特征曲线下面积在训练组和验证组分别为0.940、0.911，Hosmer⁃Lemeshow检验显示拟合良好（P>0.05），决策曲线分析证实临床适用性佳。 结论 基于pSOFA、PRISM Ⅳ、PSS评分及N/LPR构建的列线图预测模型对脓毒症患儿28 d死亡风险具有良好的预测价值。

Objective To assess the predictive value of the Pediatric Sequential Organ Failure Assessment (pSOFA), Pediatric Index of Mortality 3 (PIM-3), Pediatric Risk of Mortality IV (PRISM IV), Phoenix Sepsis Score (PSS), Phoenix Sepsis Score-8 (Phoenix-8), the ratio of neutrophil count to lymphocyte and platelet count (N/LPR), and a nomogram constructed from these factors for 28-day mortality in children with sepsis. Methods A retrospective analysis was conducted on 267 children with sepsis admitted to the Pediatric Intensive Care Unit of the Third Affiliated Hospital of Zhengzhou University from January 2017 to January 2025. Patients were categorized into a survival group (n=208) and a death group (n=59). Differences in clinical indicators between the two groups were compared. Univariate and multivariable logistic regression were used to identify independent risk factors for 28-day mortality and to construct a nomogram. Predictive performance was evaluated using receiver operating characteristic curves, calibration curves, and decision curve analysis. Results pSOFA, PIM-3, PRISM IV, PSS, Phoenix-8 scores, N/LPR, and lactate levels were significantly higher in the death group than in the survival group (P<0.05). Multivariable logistic regression identified pSOFA, PRISM IV, PSS, and N/LPR as independent risk factors for 28-day mortality. The nomogram based on these factors achieved the area under the receiver operating characteristic curves of 0.940 in the training group and 0.911 in the validation group, with good calibration according to the Hosmer-Lemeshow test (P>0.05), and decision curve analysis indicated good clinical applicability. Conclusions A nomogram incorporating pSOFA, PRISM IV, PSS, and N/LPR shows good predictive value for 28-day mortality risk in children with sepsis.