目的 探讨儿童脓毒症性凝血病（sepsis-induced coagulopathy, SIC）发生的预测因素，并建立预测模型。 方法 回顾性收集2021年7月—2023年12月北京儿童医院新疆医院重症监护病房治疗的脓毒症患儿的临床资料，按照是否出现SIC分为SIC组（n=64）和非SIC组（n=61）。采用多因素logistic回归分析筛选出SIC发生的独立预测因素。并基于这些因素建立SIC发生的预测模型，采用受试者操作特征曲线的曲线下面积、校准曲线及临床决策曲线评估模型的预测效能及临床应用价值。 结果 多因素logistic回归分析显示，降钙素原、儿童序贯器官功能衰竭评估评分及血小板平均体积是脓毒症患儿发生SIC的预测指标（P<0.05）。基于这3个指标构建的预测模型的曲线下面积为0.903（95%CI：0.852~0.953，P<0.001），其灵敏度和特异度分别为0.922、0.738。校准曲线分析表明，模型预测与实际观测结果具有良好的一致性。临床决策曲线分析表明，预测模型的临床获益良好。 结论 降钙素原、儿童序贯器官功能衰竭评估评分及血小板平均体积是儿童脓毒症发生SIC的预测指标；基于这3个预测指标建立的预测模型效能较高，具有良好的临床实用性。

Objective To identify predictors of sepsis-induced coagulopathy (SIC) in children and to establish a prediction model. Methods Clinical data were retrospectively collected from children with sepsis treated in the pediatric intensive care unit of Xinjiang Hospital of Beijing Children's Hospital between July 2021 and December 2023. Patients were classified into the SIC group (n=64) and the non-SIC group (n=61) according to whether SIC occurred. Multivariable logistic regression was employed to identify independent predictors of SIC. A prediction model was developed based on these factors. The predictive performance and clinical utility of the model were evaluated using the area under the receiver operating characteristic curve, calibration curve, and decision curve analysis. Results The multivariable logistic regression analysis showed that procalcitonin, Pediatric Sequential Organ Failure Assessment (pSOFA) score, and mean platelet volume were independent predictors of SIC in children with sepsis (P<0.05). The model developed from these three predictors yielded an area under the curve of 0.903 (95%CI: 0.852-0.953; P<0.001), with sensitivity and specificity of 0.922 and 0.738, respectively. The calibration curve analysis indicated good agreement between predicted and observed outcomes. The decision curve analysis showed favorable clinical benefit of the prediction model. Conclusions Procalcitonin, pSOFA score, and mean platelet volume are predictors of SIC among children with sepsis; the prediction model based on these three predictors shows high performance and has good clinical applicability.