Chen Zhiwei, Jia Wanyu, Guo Caili, Wu Yanwen, Liu Jian, Song Chunlan
Children's Hospital Affiliated to Zhengzhou University, Henan Children's Hospital, Zhengzhou Children's Hospital, Zhengzhou, Henan, China.
Department of Pediatrics, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China.
Pediatr Blood Cancer. 2025 Mar;72(3):e31482. doi: 10.1002/pbc.31482. Epub 2024 Dec 17.
The purpose of this study was to develop a straightforward, easy-to-use online dynamic nomogram for the identification of children who are at high risk of developing acute kidney injury (AKI) after allogeneic hematopoietic stem cell transplantation (allo-HSCT).
This was a two-center study in which 242 children in Henan Provincial Children's Hospital composed the training cohort, and 115 children in the First Affiliated Hospital of Zhengzhou University composed the validation cohort. Kaplan-Meier survival analysis was used to compare survival between children with nonacute kidney injury (NAKI) and children with AKI. Multivariate logistic regression analysis was used to identify risk factors for AKI in children who underwent HSCT. The selected variables were utilized to construct nomograms, which were validated via the concordance index (C-index), decision curve analysis, calibration curve analysis, and receiver operating characteristic (ROC) curve analysis.
Cumulative survival was significantly lower in children with AKI than in children without kidney injury (p < 0.01). Eight variables were included in the nomogram: hepatic veno-occlusive disease (HVOD), graft-versus-host disease (GVHD), ferritin, C-reactive protein (CRP), Cytomegalovirus infection (CMV), thrombotic microangiopathy (TMA), human leukocyte antigen (HLA), and nephrotoxic drugs. The nomogram calibration curves in the training and validation cohorts were highly comparable to the standard curves. The areas under the curve (AUCs) of the prediction model were 0.963 and 0.910 in the training cohort and validation cohort, respectively. The decision curve analysis (DCA) revealed that the model had a significant clinical benefit.
The occurrence of AKI affects the prognosis of children who undergo HSCT. We developed a dynamic online nomogram for predicting AKI in children who underwent allo-HSCT on the basis of eight variables. The predictive value and clinical benefit of the nomogram model were acceptable.
本研究旨在开发一种简单易用的在线动态列线图，用于识别异基因造血干细胞移植（allo-HSCT）后发生急性肾损伤（AKI）风险较高的儿童。
这是一项双中心研究，河南省儿童医院的242名儿童组成训练队列，郑州大学第一附属医院的115名儿童组成验证队列。采用Kaplan-Meier生存分析比较非急性肾损伤（NAKI）儿童和AKI儿童的生存率。多因素逻辑回归分析用于确定接受HSCT儿童发生AKI的危险因素。利用选定的变量构建列线图，并通过一致性指数（C指数）、决策曲线分析、校准曲线分析和受试者操作特征（ROC）曲线分析进行验证。
AKI儿童的累积生存率显著低于无肾损伤儿童（p<0.01）。列线图纳入了8个变量：肝静脉闭塞病（HVOD）、移植物抗宿主病（GVHD）、铁蛋白、C反应蛋白（CRP）、巨细胞病毒感染（CMV）、血栓性微血管病（TMA）、人类白细胞抗原（HLA）和肾毒性药物。训练队列和验证队列中的列线图校准曲线与标准曲线高度可比。预测模型在训练队列和验证队列中的曲线下面积（AUC）分别为0.963和0.910。决策曲线分析（DCA）显示该模型具有显著的临床益处。
AKI的发生影响接受HSCT儿童的预后。我们基于8个变量开发了一种动态在线列线图，用于预测接受allo-HSCT儿童的AKI。列线图模型的预测价值和临床益处是可以接受的。
