Zhang Xin, Li Lanfang, Zhang Chuanpeng, Zhang Huixian, Huang Haining
Department of Clinical Pharmacy, Affiliated Hospital of Jining Medical University, Jining Medical University, Jining, Shandong, China.
Medical Research Center, Affiliated Hospital of Jining Medical University, Jining Medical University, Jining, Shandong, China.
Sci Rep. 2025 Jul 2;15(1):22593. doi: 10.1038/s41598-025-07116-0.
The aim of this study was to identify risk factors more comprehensively and develop the first nomogram for predicting tigecycline-associated drug induced liver injury (DILI) in a Chinese population. Patients who underwent tigecycline treatment from January 1, 2021, to July 31, 2024, at the Affiliated Hospital of Jining Medical University were included in this retrospective study. Candidate variables were selected using least absolute shrinkage and selection operator (Lasso) regression and support vector machine recursive feature elimination (SVM-RFE), followed by univariate and multivariate logistic regression to identify independent risk factors, which were visualized in a nomogram. Nomogram model performance was evaluated via the area under the receiver operating characteristic curve (AUC). A total of 357 patients were enrolled, including 73 patients (20.4%) diagnosed with DILI, and 284 patients (79.6%) without. Fourteen intersected variables were screened through Lasso regression and SVM-RFE method. Seven variables were identified as independent risk factors and were used to construct prediction nomogram model. The AUC values of 0.82 (95% CI: 0.76-0.88) in the training cohort and 0.80 (95% CI: 0.70-0.89) indicated the nomogram model had satisfactory prediction ability. Furthermore, decision curve analysis (DCA) revealed that the nomogram provided a significant net benefit in the identifying patients at high risk of tigecycline-associated DILI. This study was the first to identify patients treated with voriconazole, with a history of malignant tumors, in a state of septic shock, and with intra-abdominal infections were at a significantly elevated risk of developing tigecycline-associated DILI. The constructed nomogram demonstrated a high level of accuracy, showcasing substantial potential to aid clinicians in pinpointing risk factors and implementing preventive measures.
本研究的目的是更全面地识别危险因素，并开发首个用于预测中国人群中替加环素相关药物性肝损伤（DILI）的列线图。本回顾性研究纳入了2021年1月1日至2024年7月31日在济宁医学院附属医院接受替加环素治疗的患者。使用最小绝对收缩和选择算子（Lasso）回归以及支持向量机递归特征消除（SVM-RFE）选择候选变量，随后进行单因素和多因素逻辑回归以识别独立危险因素，并将其在列线图中可视化。通过受试者操作特征曲线（AUC）下的面积评估列线图模型性能。共纳入357例患者，其中73例（20.4%）被诊断为DILI，284例（79.6%）未患DILI。通过Lasso回归和SVM-RFE方法筛选出14个交叉变量。7个变量被确定为独立危险因素，并用于构建预测列线图模型。训练队列中的AUC值为0.82（95%CI：0.76 - 0.88），验证队列中的AUC值为0.80（95%CI：0.70 - 0.89），表明列线图模型具有令人满意的预测能力。此外，决策曲线分析（DCA）显示，该列线图在识别替加环素相关DILI高危患者方面提供了显著的净效益。本研究首次发现接受伏立康唑治疗、有恶性肿瘤病史、处于感染性休克状态且有腹腔内感染的患者发生替加环素相关DILI的风险显著升高。构建的列线图显示出高度的准确性，在帮助临床医生确定危险因素和实施预防措施方面具有巨大潜力。
