替加环素致血小板减少症风险预测模型的建立与验证

doi:10.5246/jcps.2026.05.035

中国药学(英文版) ›› 2026, Vol. 35 ›› Issue (5): 491-501.DOI: 10.5246/jcps.2026.05.035

替加环素致血小板减少症风险预测模型的建立与验证

陈海英¹^,², 吴诗雅²^,³, 鞠建杰³, 周菲菲⁴, 林琦²^,³^,^*()

^1. 莆田平民医院，福建莆田 351100
^2. 莆田学院附属医院药学部，福建莆田 351100
^3. 福建医科大学药学院，福建福州 350000
^4. 莆田市中医院，福建莆田 351100

收稿日期:2026-01-16 修回日期:2026-02-21 接受日期:2026-03-23 出版日期:2026-05-31 发布日期:2026-05-31
通讯作者: 林琦

Development and validation of a risk prediction model for tigecycline-induced thrombocytopenia

Haiying Chen¹^,², Shiya Wu²^,³, Jianjie Ju³, Feifei Zhou⁴, Qi Lin²^,³^,^*()

^1. Putian Pingmin Hospital, Putian 351100, Fujian, China
^2. Department of Pharmacy, the Affiliated Hospital of Putian University, Putian 351100, Fujian, China
^3. School of Pharmacy, Fujian Medical University, Fuzhou 350000, Fujian, China
^4. Putian Hospital of Traditional Chinese Medicine, Putian 351100, Fujian, China

Received:2026-01-16 Revised:2026-02-21 Accepted:2026-03-23 Online:2026-05-31 Published:2026-05-31
Contact: Qi Lin
Supported by:
The Fujian Provincial Natural Science Foundation of China (Grant No. 2024J011468), the Putian City Joint Fund for Scientific and Technological Innovation in Healthcare, China (Grant No. 2024SJYL044), and the Medical Research Foundation of Putian University (Grant No. 2024107).

摘要/Abstract

摘要：

本研究旨在通过构建并验证替加环素(Tigecycline, TGC)致血小板减少症的风险预测模型，识别关键危险因素，为临床早期预警提供工具。研究回顾性收集2020年1月至2024年1月某三甲医院使用TGC的住院患者资料，采用Boruta和最小绝对收缩与选择算子(LASSO)对64个临床特征进行筛选，通过10折交叉验证、受试者工作特征曲线和校准曲线等评估模型性能，并开发动态列线图。TGC致血小板减少症研究队列共纳入919例，其中发生血小板减少症224例，发生率为24.37%，LASSO-Boruta算法分析显示年龄、入住ICU、机械辅助通气、维持剂量、基线血小板计数、红细胞计数、肌酐、钾离子、尿素氮是关键预测因子。预测模型的曲线下面积(AUC)为训练集0.744，测试集0.736，校准曲线的Brier评分为0.149。决策曲线分析显示在18%–63%的阈值概率范围内具有临床净获益。本研究构建的logistic回归模型和动态列线图具有良好的区分度和校准度，可为临床早期识别高危患者提供参考。

关键词: 替加环素, 血小板减少症, LASSO-Boruta算法, 风险因素, 预测模型

Abstract:

The study aimed to construct and validate a logistic regression–based prediction model for tigecycline (TGC)–induced thrombocytopenia, to delineate its principal risk determinants, and to develop a clinically applicable tool for early risk stratification. A retrospective cohort study was performed using data from hospitalized patients who received TGC therapy at a tertiary medical center between January 2020 and January 2024. A total of 64 candidate clinical variables were initially considered and subsequently selected using a hybrid LASSO-Boruta algorithm. Model performance was assessed through 10-fold cross-validation, receiver operating characteristic (ROC) analysis, and calibration curve evaluation. A dynamic nomogram was further developed to facilitate bedside implementation. Of the 919 eligible patients, 224 experienced thrombocytopenia, corresponding to an incidence rate of 24.37%. Nine variables emerged as independent predictors, including age, intensive care unit admission, mechanical ventilation, maintenance dose of TGC, baseline platelet count, red blood cell count, serum creatinine, potassium, and blood urea nitrogen. The model achieved an area under the curve (AUC) of 0.744 in the training cohort and 0.736 in the validation cohort, with a Brier score of 0.149, indicating favorable discriminative ability and calibration. Decision curve analysis (DCA) demonstrated a meaningful clinical net benefit across threshold probabilities ranging from 18% to 63%. Overall, the proposed logistic regression model and its accompanying dynamic nomogram exhibited robust predictive performance and clinical utility, offering a practical and interpretable approach for the early identification of patients at elevated risk of TGC-induced thrombocytopenia.

Key words: Tigecycline, Thrombocytopenia, LASSO-Boruta algorithm, Risk factors, Prediction model

Supporting: /attached/file/20260603/20260603160632_404.pdf

陈海英, 吴诗雅, 鞠建杰, 周菲菲, 林琦. 替加环素致血小板减少症风险预测模型的建立与验证[J]. 中国药学(英文版), 2026, 35(5): 491-501.

Haiying Chen, Shiya Wu, Jianjie Ju, Feifei Zhou, Qi Lin. Development and validation of a risk prediction model for tigecycline-induced thrombocytopenia[J]. Journal of Chinese Pharmaceutical Sciences, 2026, 35(5): 491-501.

图/表 5

Table 1. Univariate analysis. This table provides a comparison between the variables used in the training set.

Figure 1. illustrates the selection of salient features by the LASSO and Boruta algorithms in the training dataset. (A) Factor screening using the LASSO regression model. The left and right dashed vertical lines represent the optimal lambda (λmin) and the lambda with performance within one standard error of the optimum (λ1se), respectively; (B) Variable coefficient trajectories from the LASSO regression; (C) Feature selection using the Boruta algorithm; (D) Common predictors identified by both the Boruta and LASSO methods.

Figure 2. Establishment and assessment of the logistic regression model for thrombocytopenia. (A) ROC curve for the training set; (B) ROC curve for the test set; (C) DCA for the test set; (D) Calibration curve.

Table 2. Risk factors and their parameters of the logistic model.

Figure 3. Dynamic nomogram for predicting thrombocytopenia.

参考文献 26

[1]	Mohamed, R.A.E.; Moustafa, N.M.; Mahmoud, F.M.; Elsaadawy, Y.S.; Abdel Aziz, H.S.; Abou Bakr Gaber, S.; Hussin, A.M.; Seadawy, M.G. Whole-genome sequencing of two multidrug-resistant acinetobacter baumannii strains isolated from a neonatal intensive care unit in Egypt: a prospective cross-sectional study. BMC Microbiol. 2024, 24, 362.
[2]	Korczak, L.; Majewski, P.; Rombel, K.; Iwaniuk, D.; Sacha, P.; Modzelewski, M.; Tryniszewska, E. In vitro evaluation of tigecycline synergy testing with nine antimicrobial agents against Enterobacter cloacae clinical strains. Front. Microbiol. 2024, 15, 1490032.
[3]	Fan, X.Y.; Jiang, Y.; Wu, H.; Liu, J.; Gu, Q.Y.; Wang, Z.Y.; Sun, L.; Jiao, X.N.; Li, Q.C.; Wang, J. Distribution and spread of tigecycline resistance gene Tet (X4) in Escherichia coli from different sources. Front. Cell Infect. Microbiol. 2024, 14, 1399732.
[4]	Cui, N.N.; Cai, H.L.; Li, Z.T.; Lu, Y.T.; Wang, G.B.; Lu, A.W. Tigecycline-induced coagulopathy: a literature review. Int. J. Clin. Pharm. 2019, 41, 1408–1413.
[5]	Li, Z.K.; Zheng, P. Acute pancreatitis caused by tigecycline and furosemide combination treatment and hypofibrinogenemia caused by tigecycline: a case report. Int. J. Clin. Pharmacol. Ther. 2023, 61, 466–470.
[6]	Wu, S.Y.; Chen, Y.H.; Fan, W.J.; Wu, X.R.; Zhang, C.F.; Lin, Y.C.; Lin, Q. Predicting tigecycline-related adverse events in infected patients: a machine learning approach with clinical interpretability. Front. Pharmacol. 2025, 16, 1697929.
[7]	Leng, B.; Xue, Y.C.; Zhang, W.; Gao, T.T.; Yan, G.Q.; Tang, H. A retrospective analysis of the effect of tigecycline on coagulation function. Chem. Pharm. Bull. 2019, 67, 258–264.
[8]	Guo, M.X.; Liang, J.W.; Li, D.D.; Zhao, Y.; Xu, W.Y.; Wang, L.; Cui, X.L. Coagulation dysfunction events associated with tigecycline: a real-world study from FDA adverse event reporting system (FAERS) database. Thromb. J. 2022, 20, 12.
[9]	Li, M.X.; He, J.; Dong, G.Q.; Hu, L.L.; Shao, H. Serum concentration threshold and risk factors of tigecycline-induced hypofibrinogenaemia in critically ill patients. J. Antimicrob. Chemother. 2025, 80, 200–208.
[10]	Ma, C.Q.; Ren, X.L.; Pang, N.; Liu, Y.K.; Chen, M.F.; Zhang, X.H.; Liu, Y.; Huang, L. Incidence, characteristics, and risk factors of hypofibrinogenemia induced by generic tigecycline: a retrospective study. Naunyn Schmiedeberg’s Arch. Pharmacol. 2025, 398, 2717–2727.
[11]	Zhang, L.; Cai, X.F.; Peng, F.C.; Tian, S.S.; Wu, X.J.; Li, Y.; Guo, J.L. Comparison of bleeding risk and hypofibrinogenemia-associated risk factors between tigecycline with cefoperazone/sulbactam therapy and other tigecycline-based combination therapies. Front. Pharmacol. 2023, 14, 1182644.
[12]	Van Dung, H.; Tan, V.M.; Dieu, N.T.; Van Linh, P.; Van Khai, N.; Ngan, T.T.; Phuong, N.T.T. Development and external validation of a machine learning model for predicting drug-induced immune thrombocytopenia in a real-world hospital cohort. BMC Med. Inform. Decis. Mak. 2025, 25, 265.
[13]	Clerkin, K.J.; Salako, O.; Fried, J.A.; Griffin, J.M.; Raikhelkar, J.; Jain, R.; Restaino, S.; Colombo, P.C.; Takeda, K.; Farr, M.A.; Sayer, G.; Uriel, N.; Topkara, V.K. Impact of temporary percutaneous mechanical circulatory support before transplantation in the 2018 heart allocation system. JACC Heart Fail. 2022, 10, 12–23.
[14]	Liu, X.Z.; Duan, M.J.; Huang, H.D.; Zhang,Y.; Xiang, T.Y.; Niu, W.C.; Zhou, B.; Wang, H.L.; Zhang, T.T. Predicting diabetic kidney disease for type 2 diabetes mellitus by machine learning in the real world: a multicenter retrospective study. Front. Endocrinol. 2023, 14, 1184190.
[15]	Sun, T.; Liu, J.; Yuan, H.Q.; Li, X.; Yan, H. Construction of a risk prediction model for lung infection after chemotherapy in lung cancer patients based on the machine learning algorithm. Front. Oncol. 2024, 14, 1403392.
[16]	Zhu, Y.C.; Zhao, F.; Jin, P.F. Clinical manifestations and risk factors of tigecycline-associated thrombocytopenia. Infect. Drug Resist. 2023, 16, 6225–6235.
[17]	Wan, Q.Q.; Gao, F.; Ye, Q.F.; Wan, Q.Q.; Liu, S.; Zhou, J.D. Distribution and resistance of pathogens in liver transplant recipients with Acinetobacter baumannii infection. Ther. Clin. Risk Manag. 2015, 501–505.
[18]	Singh, A.; Uzun, G.; Bakchoul, T. Primary immune thrombocytopenia: novel insights into pathophysiology and disease management. J. Clin. Med. 2021, 10, 789.
[19]	Du, Y.J.; Yu, L.X.; Deng, B.; Li, Q.Y.; Hu, J.R.; Li, L.J.; Xu, Y.S.; Song, L.W.; Xie, F.; Wang, Y.H.; Chen, Y.H.; Liu, C.X.; Zhai, X.J.; Lu, Y.N. Pharmacokinetic interactions between tegoprazan and the combination of clarithromycin, amoxicillin and bismuth in healthy Chinese subjects: an open-label, single-center, multiple-dosage, self-controlled, phase I trial. Clin. Drug Investig. 2024, 44, 343–355.
[20]	Vaiyapuri, S.; Ali, M.S.; Moraes, L.A.; Sage, T.Y.; Lewis, K.R.; Jones, C.I.; Gibbins, J.M. Tangeretin regulates platelet function through inhibition of phosphoinositide 3-kinase and cyclic nucleotide signaling. Arterioscler. Thromb. Vasc. Biol. 2013, 33, 2740–2749.
[21]	Fantin, F.; Manica, A.; Soldani, F.; Bissoli, L.; Zivelonghi, A.; Zamboni, M. Use of tigecycline in elderly patients for Clostridium difficile infection. Geriatr. Gerontol. Int. 2015, 15, 230–231.
[22]	Han, P.P.; Yu, T.S.; Hou, Y.; Zhao, Y.J.; Liu, Y.; Sun, Y.Q.; Wang, H.Y.; Xu, P.C.; Li, G.S.; Sun, T.; Hu, X.; Liu, X.G.; Li, L.Z.; Peng, J.; Zhou, H.; Hou, M. Low-dose decitabine inhibits cytotoxic T lymphocytes-mediated platelet destruction via modulating PD-1 methylation in immune thrombocytopenia. Front. Immunol. 2021, 12, 630693.
[23]	Moming, A.; Shi, S.; Shen, S.; Qiao, J.; Yue, X.H.; Wang, B.; Ding, J.T.; Hu, Z.H.; Deng, F.; Zhang, Y.J.; Sun, S.R. Fine mapping epitope on glycoprotein-gn from severe fever with thrombocytopenia syndrome virus. PLoS One. 2021, 16, e0248005.
[24]	Liu, J.W.; Chu, M.; Jiao, Y.J.; Zhou, C.M.; Qi, R.; Yu, X.J. SFTSV infection induced interleukin-1β secretion through NLRP3 inflammasome activation. Front. Immunol. 2021, 12, 595140.
[25]	Liu, X.Z.; Liu, J.J.; Xue, W.J.; Xing, H.Y.; Zhao, L.Y.; Fang, X. Retrospective analysis of tigecycline-associated hypofibrinogenemia in patients with hypoproteinemia. J. Chin. Pharm. Sci. 2025, 34, 581–591.
[26]	Shen, L.; Fang, W.; Zhang, Y.; Pan, Y.; Zhou, Z.W. Retrospective analysis of coagulopathy in patients with severe renal insufficiency caused by tigecycline. China Pharm. 2023, 34, 2766–2769.

替加环素致血小板减少症风险预测模型的建立与验证

Development and validation of a risk prediction model for tigecycline-induced thrombocytopenia

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 5

参考文献 26

相关文章 2

编辑推荐

Metrics

本文评价

[1]	刘秀珍, 刘建军, 薛未菊, 邢海燕, 赵营莉, 方兴. 低蛋白血症患者替加环素相关低纤维蛋白原血症的真实世界研究[J]. 中国药学(英文版), 2025, 34(6): 581-591.
[2]	杨晗春, 成斐, 黄娟娟. 艾曲波帕对难治性原发免疫性血小板减少症的疗效及其对细胞免疫功能的影响[J]. 中国药学(英文版), 2023, 32(9): 736-743.