Abstract:

Objective　To explore the risk factors for cardiac damage in children with Kawasaki disease (KD), and to identify cardiac damage as early as possible and make timely intervention measures. Methods　The clinical data of 437 children with KD in the First Affiliated Hospital of Zhengzhou University from May 2018 to November 2023 were collected for a retrospective study, including 249 boys and 188 girls, aged from 2 months to 10 years. The patients were divided into a cardiac damage group (52 cases) and a non-cardiac damage group (385 cases) according to whether cardiac damage occurred. The clinical data of the two groups were compared, the risk factors of cardiac damage were analyzed, and the predictive value of the logistic prediction model was analyzed using receiver operating characteristic curve (ROC). Independent sample t test and χ² test were used. Results　The proportions of the children with fever duration ≥10 d, intravenous immunoglobulin (IVIG) treatment delay, IVIG non-response, white blood cell count ≥10×10⁹/L, total bile acid >14 μmol/L, C-reactive protein (CRP) >46 mg/L, and hemoglobin (HGB) ≥90 g/L in the cardiac damage group were higher than those in the non-cardiac damage group (all P<0.05). The duration of fever, IVIG treatment delay, IVIG non-response, CRP, and HGB were the risk factors for cardiac damage in children with KD (all P<0.05); Logit (P) =-11.754+1.333× duration of fever +1.520×IVIG treatment delay +1.761×IVIG non-response +1.676×CRP+1.345×HGB. The logistic regression equation predicted cardiac damage with an area under the curve (AUC) of 0.905 (95%CI: 0.874-0.931), with an optimal sensitivity and specificity of 90.38% and 90.65%, respectively, when compared with each of the correlates alone. Conclusions　The occurrence of cardiac damage in children with KD is mainly related to the duration of fever, IVIG treatment delay, IVIG non-response, CRP, HGB, etc. Timely treatment and more strict monitoring and follow-up should be given to the children with the above clinical characteristics to optimize the long-term prognosis.

Key words:

Kawasaki disease, Cardiac damage, Risk factors, Prediction model, Children

摘要：

目的　探讨川崎病（KD）患儿发生心脏损害的危险因素，旨在尽早识别心脏损害并及时做出干预措施。方法　收集2018年5月至2023年11月郑州大学第一附属医院437例KD患儿临床资料进行回顾性研究，其中男249例，女188例，年龄2个月～10岁。根据是否发生心脏损害分为心脏损害组（52例）和非心脏损害组（385例），比较两组临床资料，分析心脏损伤发生的危险因素，利用受试者操作特征曲线（ROC）分析logistic预测模型的预测价值。采用独立样本t检验、χ²检验。结果　心脏损害组发热持续时间≥10 d、静脉注射免疫球蛋白（IVIG）治疗延迟、IVIG无反应、白细胞计数≥10×10⁹/L、总胆汁酸>14 μmol/L、C反应蛋白（CRP）>46 mg/L、血红蛋白（HGB）≥90 g/L患儿比例高于非心脏损害组（均P<0.05）。发热持续时间、IVIG治疗延迟、IVIG无反应、CRP、HGB是KD患儿发生心脏损害的危险因素（均P<0.05）；Logit（P）=-11.754+1.333×发热持续时间+1.520×IVIG治疗延迟+1.761×IVIG无反应+1.676×CRP+1.345×HGB。与单独的各相关因素比较，logistic回归方程预测心脏损害的曲线下面积（AUC）为0.905（95%CI：0.874～0.931），最佳灵敏度与特异度分别为90.38%、90.65%。结论　KD患儿发生心脏损害主要与发热持续时间、IVIG治疗延迟、IVIG无反应、CRP、HGB等有关，对于具备上述临床特征患儿，及时给予治疗、更严格地监测和随访，以优化远期预后。

关键词:

川崎病, 心脏损害, 危险因素, 预测模型, 小儿