School of Nursing, Shanghai Jiao Tong University, No. 227 Chongqing South Road, Shanghai, China.
College of Clinical Medicine for Obstetrics & Gynecology and Pediatrics, Fujian Medical University, Fujian Children's Hospital (Fujian Branch of Shanghai Children's Medical Center), No. 966, Hengyu Road, Jin'an District, Fuzhou, China.
Pediatr Cardiol. 2024 Jan;45(1):68-80. doi: 10.1007/s00246-023-03297-5. Epub 2023 Sep 23.
Delirium is a common postoperative complication in children with congenital heart disease, which affects their postoperative recovery. The purpose of this study is to explore the risk factors of delirium and construct a nomogram model to provide novel references for the prevention and management of postoperative delirium in children with congenital heart disease. 470 children after congenital heart surgery treated in the cardiac intensive care unit (CICU) of Shanghai Children's Medical Center were divided into a model and a validation cohort according to the principle of 7:3 distribution temporally. Then, the delirium-related influencing factors of 330 children in the training cohort were analyzed, and the nomogram model was established by a combination of Lasso regression and logistic regression. The data of 140 children in the validation cohort were used to verify the effectiveness of the model. Multivariable logistic regression analysis showed that age, disease severity, non-invasive ventilation after extubation, delayed chest closure, phenobarbital dosage, promethazine dosage, mannitol usage, and elevated temperature were independent risk factors for postoperative delirium. The area under the receiver operating characteristic curve (AUC) of the nomogram model was 0.864 and the Brier value was 0.121. Regarding the validation of the model's effect, our results showed that 51 cases were predicted by the model and 34 cases actually occurred, including 4 cases of false negative and 21 cases of false positive. The positive predictive value of the model was 58.8%, and its negative predictive value was 95.5%. The nomogram model established in this study showed acceptable performance in predicting postoperative delirium in children with congenital heart disease.
术后谵妄是儿童先天性心脏病的常见并发症，影响其术后恢复。本研究旨在探讨术后谵妄的危险因素，并构建列线图模型，为儿童先天性心脏病术后谵妄的预防和管理提供新的参考。将上海儿童医学中心心脏重症监护病房（CICU）收治的 470 例先天性心脏病手术后患儿，按照时间上 7:3 的原则分为模型组和验证组。然后，对训练组 330 例患儿的谵妄相关影响因素进行分析，并采用 Lasso 回归和逻辑回归相结合的方法构建列线图模型。验证组 140 例患儿的数据用于验证模型的有效性。多变量逻辑回归分析表明，年龄、疾病严重程度、拔管后无创通气、延迟关胸、苯巴比妥剂量、异丙嗪剂量、甘露醇使用和发热是术后谵妄的独立危险因素。列线图模型的受试者工作特征曲线下面积（AUC）为 0.864，Brier 值为 0.121。关于模型效果的验证，结果显示，模型预测 51 例，实际发生 34 例，包括 4 例假阴性和 21 例假阳性。模型的阳性预测值为 58.8%，阴性预测值为 95.5%。本研究建立的列线图模型在预测儿童先天性心脏病术后谵妄方面具有良好的性能。
