| 摘要: |
| 目的:构建川崎病(KD)静脉注射免疫球蛋白(IVIG)无反应性的风险预测模型,并探讨防治策略。方法:回顾南京市江
宁医院384 例KD 患儿临床资料,根据IVIG 疗效分为有反应性组(334 例)、无反应性组(50 例),分析影响IVIG 无反应性的危
险因素,构建风险预测模型并验证预测效能,绘制决策分析曲线(DCA)验证临床实用性。结果:与有反应性组比较,无反应性组
首次IVIG 治疗时间更早,红细胞沉降率(ESR)、N 末端脑钠肽(NT-proBNP) 水平、血小板淋巴细胞比值(PLR) 和系统免疫炎症
反应指数( SII) 更高( P <0. 05);logistic 回归分析显示,高水平ESR、NT-proBNP、PLR、SII 是IVIG 无反应性的危险因素( P <
0. 05);校准曲线显示,模型拟合度好(P>0. 187)。受试者工作特征(ROC)曲线结果显示,该预测模型预测IVIG 无反应性的曲
线下面积(AUC)高于Kobayashi 评分预测IVIG 无反应性的AUC(P<0. 05)。DCA 显示,该模型具有较好的获益。结论:高水平
ESR、NT-proBNP、PLR、SII 是KD 患儿IVIG 无反应性的危险因素,构建的风险预测模型具有良好的区分度、准确度和临床实用性。 |
| 关键词: 川崎病 免疫球蛋白 无反应性 血小板淋巴细胞比值 系统免疫炎症反应指数 |
| DOI：10.13407/j.cnki.jpp.1672-108X.2025.12.007 |
|
| 基金项目:江苏省卫生健康委科研项目,编号M2020195。 |
|
| Construction of High-Risk Factor Model for Non-Response to Intravenous Immunoglobulin in KawasakiDisease and Research on Prevention and Treatment Strategies |
| Shao Junjia1, Qin Xuan1, Liu Lingzhen1, Li Danxue2 |
| (1. The Affiliated Jiangning Hospital of Nanjing Medical University,
Nanjing 211100, China; 2. Xuzhou Medical University Affiliated Hospital, Jiangsu Xuzhou 221006, China) |
| Abstract: |
| Objective: To develop a risk prediction model for non-response to intravenous immunoglobulin (IVIG) in children with
Kawasaki disease (KD), and to explore preventive and therapeutic strategies. Methods: By reviewing the clinical data of 384 children
with KD in Nanjing Jiangning Hospital, all patients were divided into the responsive group (334 cases) and non-responsive group (50
cases) based on the efficacy of IVIG. The risk factors influencing the non-response to IVIG were analyzed, the risk prediction model was
constructed and its predictive efficacy was verified. The decision curve analysis (DCA) was drawn to validate the clinical practicability.
Results: Compared with the responsive group, the non-responsive group had a earlier time for the first IVIG treatment, higher levels of
erythrocyte sedimentation rate (ESR), N-terminal brain natriuretic peptide (NT-proBNP), platelet-to-lymphocyte ratio (PLR), and
systemic immune inflammatory (SII) response index (P<0. 05). Multivariate logistic regression analysis confirmed that elevated ESR,
NT-proBNP, PLR, and SII levels were independent risk factors for non-response to IVIG ( P < 0. 05). The calibration curve
demonstrated good model fit (P>0. 187). The receiver operating characteristic (ROC) curve results indicated that the area under the
curve (AUC) of the prediction model for predicting non-response to IVIG was higher than that of Kobayashi score (P<0. 05). The DCA
showed that this model had good patient benefits. Conclusion: Elevated ESR, NT-proBNP, PLR, and SII levels are risk factors for nonresponse
to IVIG in KD. The proposed risk prediction model exhibits strong discriminative ability, accuracy, and clinical utility. |
| Key words: Kawasaki disease immunoglobulin non-response platelet-lymphocyte ratio systemic immune inflammatory response index |
