Machine learning-based prediction model for hypofibrinogenemia after tigecycline therapy.
Hypofibrinogenemia
Influencing factors
Machine learning
Prediction models
Survival model
Tigecycline
Journal
BMC medical informatics and decision making
ISSN: 1472-6947
Titre abrégé: BMC Med Inform Decis Mak
Pays: England
ID NLM: 101088682
Informations de publication
Date de publication:
04 Oct 2024
04 Oct 2024
Historique:
received:
22
11
2023
accepted:
25
09
2024
medline:
5
10
2024
pubmed:
5
10
2024
entrez:
4
10
2024
Statut:
epublish
Résumé
In clinical practice, the incidence of hypofibrinogenemia (HF) after tigecycline (TGC) treatment significantly exceeds the probability claimed by drug manufacturers. We aimed to identify the risk factors for TGC-associated HF and develop prediction and survival models for TGC-associated HF and the timing of TGC-associated HF. This single-center retrospective cohort study included 222 patients who were prescribed TGC. First, we used binary logistic regression to screen the independent factors influencing TGC-associated HF, which were used as predictors to train the extreme gradient boosting (XGBoost) model. Receiver operating characteristic curve (ROC), calibration curve, decision curve analysis (DCA), and clinical impact curve analysis (CICA) were used to evaluate the performance of the model in the verification cohort. Subsequently, we conducted survival analysis using the random survival forest (RSF) algorithm. A consistency index (C-index) was used to evaluate the accuracy of the RSF model in the verification cohort. Binary logistic regression identified nine independent factors influencing TGC-associated HF, and the XGBoost model was constructed using these nine predictors. The ROC and calibration curves showed that the model had good discrimination (areas under the ROC curves (AUC) = 0.792 [95% confidence interval (CI), 0.668-0.915]) and calibration ability. In addition, DCA and CICA demonstrated good clinical practicability of this model. Notably, the RSF model showed good accuracy (C-index = 0.746 [95%CI, 0.652-0.820]) in the verification cohort. Stratifying patients treated with TGC based on the RSF model revealed a statistically significant difference in the mean survival time between the low- and high-risk groups. The XGBoost model effectively predicts the risk of TGC-associated HF, whereas the RSF model has advantages in risk stratification. These two models have significant clinical practical value, with the potential to reduce the risk of TGC therapy.
Sections du résumé
BACKGROUND
BACKGROUND
In clinical practice, the incidence of hypofibrinogenemia (HF) after tigecycline (TGC) treatment significantly exceeds the probability claimed by drug manufacturers.
OBJECTIVE
OBJECTIVE
We aimed to identify the risk factors for TGC-associated HF and develop prediction and survival models for TGC-associated HF and the timing of TGC-associated HF.
METHODS
METHODS
This single-center retrospective cohort study included 222 patients who were prescribed TGC. First, we used binary logistic regression to screen the independent factors influencing TGC-associated HF, which were used as predictors to train the extreme gradient boosting (XGBoost) model. Receiver operating characteristic curve (ROC), calibration curve, decision curve analysis (DCA), and clinical impact curve analysis (CICA) were used to evaluate the performance of the model in the verification cohort. Subsequently, we conducted survival analysis using the random survival forest (RSF) algorithm. A consistency index (C-index) was used to evaluate the accuracy of the RSF model in the verification cohort.
RESULTS
RESULTS
Binary logistic regression identified nine independent factors influencing TGC-associated HF, and the XGBoost model was constructed using these nine predictors. The ROC and calibration curves showed that the model had good discrimination (areas under the ROC curves (AUC) = 0.792 [95% confidence interval (CI), 0.668-0.915]) and calibration ability. In addition, DCA and CICA demonstrated good clinical practicability of this model. Notably, the RSF model showed good accuracy (C-index = 0.746 [95%CI, 0.652-0.820]) in the verification cohort. Stratifying patients treated with TGC based on the RSF model revealed a statistically significant difference in the mean survival time between the low- and high-risk groups.
CONCLUSIONS
CONCLUSIONS
The XGBoost model effectively predicts the risk of TGC-associated HF, whereas the RSF model has advantages in risk stratification. These two models have significant clinical practical value, with the potential to reduce the risk of TGC therapy.
Identifiants
pubmed: 39367370
doi: 10.1186/s12911-024-02694-x
pii: 10.1186/s12911-024-02694-x
doi:
Substances chimiques
Tigecycline
70JE2N95KR
Anti-Bacterial Agents
0
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
284Subventions
Organisme : Zhejiang Pharmaceutical Society Hospital Pharmacy Special Research Grant Project
ID : 2022ZYY03
Informations de copyright
© 2024. The Author(s).
Références
Magiorakos AP, Srinivasan A, Carey RB, et al. Multidrug-resistant, extensively drug-resistant and pandrug-resistant bacteria: an international expert proposal for interim standard definitions for acquired resistance. Clin Microbiol Infect. 2012;18(3):268–81.
doi: 10.1111/j.1469-0691.2011.03570.x
pubmed: 21793988
Ventola CL. The antibiotic resistance crisis: part 1: causes and threats. P t. 2015;40(4):277–83.
pubmed: 25859123
pmcid: 4378521
Rusu A, Buta EL. The Development of Third-Generation Tetracycline antibiotics and New perspectives. Pharmaceutics 2021; 13(12).
Yaghoubi S, Zekiy AO, Krutova M, et al. Tigecycline antibacterial activity, clinical effectiveness, and mechanisms and epidemiology of resistance: narrative review. Eur J Clin Microbiol Infect Dis. 2022;41(7):1003–22.
doi: 10.1007/s10096-020-04121-1
pubmed: 33403565
Dorn C, Petroff D, Kratzer A, et al. Tigecycline Soft Tissue Penetration in obese and non-obese Surgical patients determined by using in Vivo Microdialysis. Eur J Drug Metab Pharmacokinet. 2022;47(5):749–55.
doi: 10.1007/s13318-022-00789-2
pubmed: 35945479
pmcid: 9399032
Song Y, Hu L, Shu Q, et al. Tigecycline salvage therapy for critically ill children with multidrug-resistant/extensively drug-resistant infections after surgery. Int J Infect Dis. 2018;75:82–8.
doi: 10.1016/j.ijid.2018.08.007
pubmed: 30125690
Kadoyama K, Sakaeda T, Tamon A, et al. Adverse event profile of tigecycline: data mining of the public version of the U.S. Food and Drug Administration adverse event reporting system. Biol Pharm Bull. 2012;35(6):967–70.
doi: 10.1248/bpb.35.967
pubmed: 22687540
Xie W, Ma K, Xu Z, et al. Risk factors of tigecycline-associated fibrinogen reduction in patients with renal transplantation: a case-control study. Transl Androl Urol. 2022;11(10):1410–8.
doi: 10.21037/tau-22-522
pubmed: 36386261
pmcid: 9641064
Fan Q, Huang W, Weng Y, et al. Hypofibrinogenemia induced by high-dose tigecycline-case report and review of literature. Med (Baltim). 2020;99(43):e22638.
doi: 10.1097/MD.0000000000022638
Wang D, Lin C, Gu C, et al. Tigecycline-Associated Coagulopathy: a single-Center Retrospective Analysis. Pharmacology. 2022;107(9–10):524–36.
doi: 10.1159/000525380
pubmed: 35820375
Zhang Q, Wang J, Liu H, et al. Risk factors for tigecycline-induced hypofibrinogenaemia. J Clin Pharm Ther. 2020;45(6):1434–41.
doi: 10.1111/jcpt.13250
pubmed: 32860258
Kryczka KE, Kruk M, Demkow M et al. Fibrinogen and a Triad of Thrombosis, Inflammation, and the Renin-Angiotensin System in Premature Coronary Artery Disease in Women: A New Insight into Sex-Related Differences in the Pathogenesis of the Disease. Biomolecules. 2021; 11(7).
Liu J, Yan Y, Zhang F. Risk factors for Tigecycline-Associated Hypofibrinogenemia. Ther Clin Risk Manag. 2021;17:325–32.
doi: 10.2147/TCRM.S302850
pubmed: 33888987
pmcid: 8057826
Hu J, Xiao YH, Zheng Y, et al. Clinical characteristics and risk factors of tigecycline-associated hypofibrinogenaemia in critically ill patients. Eur J Clin Pharmacol. 2020;76(7):913–22.
doi: 10.1007/s00228-020-02860-w
pubmed: 32355990
pmcid: 7224009
Wu X, Zhao P, Dong L, et al. A case report of patient with severe acute cholangitis with tigecycline treatment causing coagulopathy and hypofibrinogenemia. Med (Baltim). 2017;96(49):e9124.
doi: 10.1097/MD.0000000000009124
Zhang Q, Zhou S, Zhou J. Tigecycline treatment causes a decrease in fibrinogen levels. Antimicrob Agents Chemother. 2015;59(3):1650–5.
doi: 10.1128/AAC.04305-14
pubmed: 25547356
pmcid: 4325772
Frazier, PIJapa. A tutorial on Bayesian optimization. 2018.
Ma WJJN. Bayesian decision models: A primer. 2019; 104(1): pp. 164–175.
Gao L, Ding, YJBrn. Disease prediction via bayesian hyperparameter optimization and ensemble learning. 2020; 13: pp. 1–6.
Guirao X, Sánchez García M, Bassetti M, et al. Safety and tolerability of tigecycline for the treatment of complicated skin and soft-tissue and intra-abdominal infections: an analysis based on five European observational studies. J Antimicrob Chemother. 2013;68(Suppl 2):ii37–44.
doi: 10.1093/jac/dkt143
pubmed: 23772045
Guo M, Liang J, Li D, et al. Coagulation dysfunction events associated with tigecycline: a real-world study from FDA adverse event reporting system (FAERS) database. Thromb J. 2022;20(1):12.
doi: 10.1186/s12959-022-00369-z
pubmed: 35248072
pmcid: 8898466
Pieringer H, Schmekal B, Biesenbach G, et al. Severe coagulation disorder with hypofibrinogenemia associated with the use of tigecycline. Ann Hematol. 2010;89(10):1063–4.
doi: 10.1007/s00277-010-0911-7
pubmed: 20174923
Zhang Q, Zhou J. Fibrinogenopenia caused by Tigecycline: a case report. Eur Rev Med Pharmacol Sci. 2015;19(6):915–7.
pubmed: 25855910
Wu PC, Wu CC. Tigecycline-associated hypofibrinogenemia: a case report and review of the literature. IDCases. 2018;11:56–7.
doi: 10.1016/j.idcr.2018.01.003
pubmed: 29560313
pmcid: 5857890
Sabanis N, Paschou E, Gavriilaki E, et al. Hypofibrinogenemia induced by tigecycline: a potentially life-threatening coagulation disorder. Infect Dis. 2015;47(10):743–6.
doi: 10.3109/23744235.2015.1043942
Bialkower M, Garnier G. Fibrinogen Diagnostics in Major Hemorrhage. Crit Rev Anal Chem. 2022;52(1):194–209.
doi: 10.1080/10408347.2020.1793098
pubmed: 32683907
Zhu L, Wang L, Zhang Y, et al. Fatal hemorrhagic pneumonia in patients with hematologic diseases and Stenotrophomonas maltophilia bacteremia: a retrospective study. BMC Infect Dis. 2021;21(1):723.
doi: 10.1186/s12879-021-06420-0
pubmed: 34332526
pmcid: 8325812
Kato T, Matsuura K. Recombinant human soluble thrombomodulin improves mortality in patients with sepsis especially for severe coagulopathy: a retrospective study. Thromb J. 2018;16:19.
doi: 10.1186/s12959-018-0172-6
pubmed: 30158838
pmcid: 6107946
Spoto S, Daniel Markley J, Valeriani E, et al. Active surveillance cultures and procalcitonin in Combination with Clinical data to guide empirical antimicrobial therapy in Hospitalized Medical patients with Sepsis. Front Microbiol. 2022;13:797932.
doi: 10.3389/fmicb.2022.797932
pubmed: 35464939
pmcid: 9023116
Li Z, Zeng Q, Xu S, et al. Development and validation of a Nomogram for Predicting Tigecycline-related Coagulopathy: a retrospective cohort study. Infect Drug Resist. 2023;16:423–34.
doi: 10.2147/IDR.S388438
pubmed: 36718461
pmcid: 9884007
Sulimai N, Lominadze D. Fibrinogen and Neuroinflammation during Traumatic Brain Injury. Mol Neurobiol. 2020;57(11):4692–703.
doi: 10.1007/s12035-020-02012-2
pubmed: 32776201
pmcid: 7530020
Uemura S, Higuchi R, Yazawa T, et al. Level of total bilirubin in the bile of the future remnant liver of patients with obstructive jaundice undergoing hepatectomy predicts postoperative liver failure. J Hepatobiliary Pancreat Sci. 2020;27(9):614–21.
doi: 10.1002/jhbp.784
pubmed: 32506707
Asif A, Park SH, Soomro AM, et al. Microphysiological system with continuous analysis of albumin for hepatotoxicity modeling and drug screening. J Ind Eng Chem. 2021;98:318–26.
doi: 10.1016/j.jiec.2021.03.035
Huang YT, Yu CI, Chen PY, et al. Comparison of bleeding risk between colistin-tigecycline and colistin-carbapenem treatment regimens: a retrospective cohort study. Infect Drug Resist. 2021;14:4949–55.
doi: 10.2147/IDR.S339188
pubmed: 34858035
pmcid: 8629913