Development and validation of a predictive model for peripherally inserted central catheter-related thrombosis in breast cancer patients based on artificial neural network: A prospective cohort study

https://doi.org/10.1016/j.ijnurstu.2022.104341Get rights and content

Abstract

Background

Peripherally inserted central catheters have been extensively applied in clinical practices. However, they are associated with an increased risk of thrombosis. To improve patient care, it is critical to timely identify patients at risk of developing peripherally inserted central catheter-related thrombosis. Artificial neural networks have been successfully used in many areas of clinical events prediction and affected clinical decisions and practice.

Objective

To develop and validate a novel clinical model based on artificial neural network for predicting peripherally inserted central catheter-related thrombosis in breast cancer patients who underwent chemotherapy and determine whether it may improve the prediction performance compared with the logistic regression model.

Setting

A large general hospital in Fujian Province, China.

Participants

One thousand eight hundred and forty-four breast cancer patients with peripherally inserted central catheters placement for chemotherapy were eligible for the study.

Methods

The dataset was divided into a training set (N = 1497) and an independent validation set (N = 347). The synthetic minority oversampling technique (SMOTE) was used to handle the effect of imbalance class. Both the artificial neural network and logistic regression models were then developed on the training set with and without SMOTE, respectively. The performance of each model was evaluated on the validation set using accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC).

Results

Of the 1844 enrolled patients, 256 (13.9%) were diagnosed with peripherally inserted central catheter-related thrombosis. Predictive models were constructed in the training set and assessed in the validation set. Eight factors were selected as input variables to develop the artificial neural network model. Without SMOTE, the artificial neural network model (AUC = 0.725) outperformed the logistic regression model (AUC = 0.670, p = 0.039). SMOTE improved the performance of both two models based on AUC. With the SMOTE sampling, the artificial neural network model performed the best across all evaluated models, the AUC value remained statistically better than that of the logistic regression model (0.742 vs. 0.675, p = 0.004).

Conclusion

Artificial neural network model can effectively predict peripherally inserted central catheter-related thrombosis in breast cancer patients receiving chemotherapy. Identifying high-risk groups with peripherally inserted central catheter-related thrombosis can provide close monitoring and an opportune time for intervention.

Section snippets

What is already known

  • •

    Peripherally inserted central catheters are one of the most commonly used methods for central venous catheterization. However, they are associated with an increased risk of venous thrombosis.

  • •

    Early identification of patients at risk of peripherally inserted central catheter-related thrombosis can result in early intervention and quality care.

  • •

    Up to now, no predictive model has been published to use neural networks specifically for breast cancer patients in predicting peripherally inserted central

What this paper adds

  • •

    A novel artificial neural network-based prediction model for peripherally inserted central catheter-related thrombosis in breast cancer was constructed.

  • •

    The artificial neural network model can accurately predict peripherally inserted central catheter-related thrombosis in breast cancer patients and outperform the traditional logistic regression method.

Study design and setting

This prospective cohort study was conducted on adult breast cancer patients with peripherally inserted central catheters admitted to chemotherapy between January 2018 and June 2021. All the data were collected at Fujian Medical University Union Hospital, a university-affiliated general hospital in China.

Participants

Patients were enrolled into this study if they; (a) had a histopathological diagnosis of breast cancer, (b) were ≥ 18 years of age, (c) had undergone peripherally inserted central catheters

Population characteristics

Between January 2018 and June 2021, 1844 breast cancer patients who underwent peripherally inserted central catheter insertion were finally enrolled in the study. Data collected before December 2020 were selected as the training set (n = 1497) and the remaining as the validation set (n = 347). Detailed demographic characteristics are presented in Table 1. In the study population, all patients were female and a total of 256 patients (13.9%) were diagnosed with peripherally inserted central

Discussion

The peripherally inserted central catheters are accepted worldwide as one of the most effective intravenous infusion tools. The prevalence of peripherally inserted central catheters use has increased rapidly in recent years, breast cancer patients are no exception (Chopra et al., 2013; Govindan et al., 2021; Johansson et al., 2013). However, they have been associated with an increased risk for venous thrombosis (Saber et al., 2011). Thrombosis may terminate the treatment process and even cause

Conclusion

We proposed and validated an artificial neural network-based model for the prediction of peripherally inserted central catheter-related thrombosis in breast cancer patients undergoing chemotherapy. The results indicate that the artificial neural network can accurately predict peripherally inserted central catheter-related thrombosis, and it outperformed the conventional logistic regression model. Identifying high-risk groups with peripherally inserted central catheter-related thrombosis can

Funding

This work was supported by the Joint Funds for the Innovation of Science and Technology, Fujian Province (2019Y9103), Fujian Provincial Finance Special Project (2020CZ011), and the National Key Speciality Construction Project of Clinical Nursing.

CRediT authorship contribution statement

Jianqin Fu: Methodology, Investigation, Data curation, Writing - Original draft preparation. Weifeng Cai: Methodology, Investigation, Data curation, Writing - Original draft preparation. Bangwei Zeng: Conceptualization, Methodology, Software, Formal analysis, Validation. Lijuan He: Investigation, Data Curation. Liqun Bao: Investigation, Data Curation. Zhaodi Lin: Investigation, Data Curation. Fang Lin: Investigation, Data Curation. Wenjuan Hu: Investigation, Data Curation. Linying Lin:

Declaration of Competing Interest

None.

References (50)

  • Y. Fei et al.

    Artificial neural networks predict the incidence of portosplenomesenteric venous thrombosis in patients with acute pancreatitis

    J. Thromb. Haemost.

    (2017)
  • S. Govindan et al.

    To PICC or not to PICC? A cross-sectional survey of vascular access practices in the ICU

    J. Crit. Care

    (2021)
  • M. Itkin et al.

    Peripherally inserted central catheter thrombosis–reverse tapered versus nontapered catheters: a randomized controlled study

    J. Vasc. Interv. Radiol.

    (2014)
  • M.J. Kovacs et al.

    A pilot study of central venous catheter survival in cancer patients using low-molecular-weight heparin (dalteparin) and warfarin without catheter removal for the treatment of upper extremity deep vein thrombosis (The catheter Study)

    J. Thromb. Haemost.

    (2007)
  • V. L´opez et al.

    An insight into classification with imbalanced data: empirical results and current trends on using data intrinsic characteristics

    Inform. Sci.

    (2013)
  • Y. Lin et al.

    The caprini thrombosis risk model predicts the risk of peripherally inserted central catheter-related upper extremity venous thrombosis in patients with cancer

    J. Vasc. Surg. Venous Lymphat. Disord.

    (2021)
  • S. Loibl et al.

    Breast cancer

    Lancet (London, England)

    (2021)
  • T.D. Martins et al.

    Artificial neural networks for prediction of recurrent venous thromboembolism

    Int. J. Med. Inform.

    (2020)
  • D. Periard et al.

    Randomized controlled trial of peripherally inserted central catheters vs. peripheral catheters for middle duration in-hospital intravenous therapy

    J. Thromb. Haemost.

    (2008)
  • W. Saber et al.

    Risk factors for catheter-related thrombosis (CRT) in cancer patients: a patient-level data (IPD) meta-analysis of clinical trials and prospective studies

    J. Thromb. Haemost.

    (2011)
  • B. Sousa et al.

    Central venous access in oncology: ESMO clinical practice guidelines

    Ann. Oncol.

    (2015)
  • B. Sousa et al.

    Central venous access in oncology: ESMO clinical practice guidelines

    Ann. Oncol.

    (2015)
  • S.F. Sung et al.

    Developing a stroke alert trigger for clinical decision support at emergency triage using machine learning

    Int. J. Med. Inform.

    (2021)
  • N. Vazzana et al.

    Diabetes mellitus and thrombosis

    Thromb. Res.

    (2012)
  • H.J. Baarslag et al.

    Prospective study of color duplex ultrasonography compared with contrast venography in patients suspected of having deep venous thrombosis of the upper extremities

    Ann. Intern. Med.

    (2002)
    • 1

    Jianqin Fu, Weifeng Cai and Bangwei Zeng made equal contributions to this manuscript.

    View full text
    © 2022 Elsevier Ltd. All rights reserved.