智能方法预测深颈部间隙脓肿引流术后患者入住重症监护病房：一项多中心回顾性研究。
Intelligent method to predict intensive care unit admission after drainage operation in patients with deep neck space abscess: A multicenter retrospective study.
机构信息
Otorhinolaryngology Hospital, The First Affiliated Hospital, Sun Yat-Sen University, Guangzhou, China.
Department of Otorhinolaryngology - Head and Neck Surgery, The Second Affiliated Hospital of Shantou University Medical College, Shantou, China.
出版信息
Head Neck. 2024 Oct;46(10):2484-2495. doi: 10.1002/hed.27738. Epub 2024 Mar 15.
BACKGROUNDS
A deep neck space abscess (DNSA) is a critical condition resulting from infection of deep neck fascia and soft issue, leading to high morbidity and mortality. Therefore, intensive care can be very significant for patients with DNSA. This study aimed to develop models to predict the need for postoperative intensive care in patients with DNSA.
METHODS
We retrospectively analyzed the records of 332 patients with DNSA who received drainage operation between 2015 and 2020. Multivariate logistic regression analysis and the eXtrem Gradient Boosting (XGBoost) algorithm were used to develop predictive models.
RESULTS
We developed two predictive models, the nomogram and the XGBoost model. The area under the curve (AUC) of the nomogram was 0.911 and of the XGBoost model was 0.935.
CONCLUSION
We developed two predictive models for guiding clinical decision making for postoperative ICU admission for DNSA patients, which may help improve prognosis and optimize intensive care resource allocation.
背景
深部颈间隙脓肿（DNSA）是一种严重的疾病，源于深部颈筋膜和软组织的感染，具有较高的发病率和死亡率。因此，重症监护对 DNSA 患者非常重要。本研究旨在建立预测 DNSA 患者术后需要重症监护的模型。
方法
我们回顾性分析了 2015 年至 2020 年间接受引流手术的 332 例 DNSA 患者的记录。采用多变量逻辑回归分析和极端梯度提升（XGBoost）算法建立预测模型。
结果
我们建立了两个预测模型，即列线图和 XGBoost 模型。列线图的曲线下面积（AUC）为 0.911，XGBoost 模型的 AUC 为 0.935。
结论
我们建立了两个预测模型，用于指导 DNSA 患者术后 ICU 入住的临床决策，这可能有助于改善预后和优化重症监护资源的分配。
Zotero 插件