Construction and validation of a risk prediction model for PICC related bloodstream infections in elderly acute leukemia patients during chemotherapy
10.3760/cma.j.cn115682-20230824-00678
- VernacularTitle:老年急性白血病患者化疗期间发生PICC相关血流感染风险预测模型的构建与验证
- Author:
Juan GUO
1
;
Weijie CAO
;
Suping ZHANG
;
Na LIU
;
Yihong DING
Author Information
1. 郑州大学第一附属医院血液内科，郑州 450000
- Keywords:
Leukemia;
Acute leukemia;
Elderly patients;
Peripherally inserted central catheter;
Bloodstream infections;
Prediction model
- From:
Chinese Journal of Modern Nursing
2024;30(11):1489-1496
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the risk factors for peripherally inserted central catheter (PICC) related bloodstream infections in elderly acute leukemia patients during chemotherapy, and construct and validate relevant predictive models.Methods:This study selected 291 patients from the First Affiliated Hospital of Zhengzhou University from January 2019 to December 2020 as the modeling group, and 277 patients from January 2021 to December 2022 as the validation group. Multivariate Logistic regression was used to determine the risk factors for PICC related bloodstream infections during chemotherapy in elderly acute leukemia patients, and a prediction model was constructed and validated based on the analysis results.Results:The incidence of PICC related bloodstream infections in the modeling group was 8.2% (24/291). Multivariate Logistic regression analysis showed that chemotherapy frequency, single catheterization puncture frequency, whether catheterization maintenance frequency was standardized, and catheterization retention time were risk factors for PICC related bloodstream infections in elderly acute leukemia patients during chemotherapy. Based on this result, a prediction model was constructed, and the area under the receiver operating characteristic curve ( AUC) of the modeling group's prediction model was 0.798 [95% CI (0.734, 0.869) ], while the AUC of the validation group's prediction model was 0.745 [95% CI (0.712, 0.844) ]. The calibration curves of the modeling and validation groups showed that the prediction model had high predictive performance. Conclusions:A prediction model for PICC related bloodstream infections in elderly acute leukemia patients during chemotherapy is constructed based on the results of multiple factor analysis, and the predictive performance of the model is verified. Nursing staff can quantify the risk of PICC related bloodstream infections in patients based on this prediction model and implement targeted nursing measures.
