Zheng Zihao, Tang Siyu, Shao Ziqiang, Cai Hanhui, Wang Jiangbo, Lu Lihai, Yang Xianghong, Liu Jingquan
Emergency and Critical Care Center, Intensive Care Unit, Zhejiang Provincial People's Hospital, Affiliated People's Hospital, Hangzhou Medical College, Hangzhou, 310014, Zhejiang Province, China.
The Fourth School of Clinical Medicine, Zhejiang Chinese Medical University, Hangzhou, 310053, Zhejiang Province, China.
Heliyon. 2024 Sep 5;10(17):e37498. doi: 10.1016/j.heliyon.2024.e37498. eCollection 2024 Sep 15.
Enteral nutrition administered via the nasointestinal tube (NET) is a prevalent nutritional modality among critically ill patients, and abdominal radiographs hold significant value in accurately ascertaining the precise positioning of the NET subsequent to its placement. Therefore, we propose an innovative approach to construct a clinical prediction model based on NET's configuration within the gastrointestinal tract in abdominal radiography. This model aims to enhance the accuracy of determining the position of NETs after their placement.
Patients admitted to the intensive care unit of Zhejiang Provincial People's Hospital between October 2017 and October 2021 were included to constitute the training cohort for retrospective analysis, and nomogram was constructed. Consecutively enrolled patients admitted to the same hospital from October 2021 to October 2023 were included as the validation cohort. The training cohort underwent a univariate analysis initially, followed by a multivariate logistic regression approach to analyze and identify the most appropriate model. Subsequently, nomogram was generated along with receiver operator characteristic curves, calibration curves, and decision curves for both the training and validation cohorts to evaluate the predictive performance of the model.
The training and validation cohorts comprised 574 and 249 patients, respectively, with successful tube placement observed in 60.1 % and 76.3 % of patients, correspondingly. The predictors incorporated in the prediction maps encompass the "C-shape," the height of "inverse C-shape," showing the duodenojejunal flexure, and the location of the head end of the NET. The model demonstrated excellent predictive efficacy, achieving an AUC of 0.883 (95 % CI 0.855-0.911) and good calibration. Furthermore, when applied to the validation cohort, the nomogram exhibited strong discrimination with an AUC of 0.815 (95 % CI 0.750-0.880) and good calibration.
The combination of abdominal radiography and NET's configuration within the gastrointestinal tract enables accurate determination of NET placement in critically ill patients.
经鼻肠管（NET）给予肠内营养是重症患者中普遍采用的营养方式，腹部X线片对于准确确定NET放置后的精确位置具有重要价值。因此，我们提出一种创新方法，基于腹部X线片中NET在胃肠道内的形态构建临床预测模型。该模型旨在提高NET放置后确定其位置的准确性。
纳入2017年10月至2021年10月在浙江省人民医院重症监护病房住院的患者作为训练队列进行回顾性分析，并构建列线图。将2021年10月至2023年10月在同一家医院连续纳入的患者作为验证队列。训练队列首先进行单因素分析，然后采用多因素逻辑回归方法进行分析，以确定最合适的模型。随后，为训练队列和验证队列生成列线图以及受试者操作特征曲线、校准曲线和决策曲线，以评估模型的预测性能。
训练队列和验证队列分别包括574例和249例患者，相应地，成功置管的患者分别为60.1%和76.3%。预测图中纳入的预测因素包括“C形”、显示十二指肠空肠曲的“倒C形”高度以及NET头端的位置。该模型显示出优异的预测效能，曲线下面积（AUC）为0.883（95%置信区间0.855 - 0.911），校准良好。此外，当应用于验证队列时，列线图显示出较强的区分能力，AUC为0.815（95%置信区间0.750 - 0.880），校准良好。
腹部X线片与NET在胃肠道内的形态相结合，能够准确确定重症患者的NET放置位置。
