Skip to main content

Advertisement

SpringerLink
Log in

Deep learning model for prediction of extended-spectrum beta-lactamase (ESBL) production in community-onset Enterobacteriaceae bacteraemia from a high ESBL prevalence multi-centre cohort

  • Original Article
  • Published:
European Journal of Clinical Microbiology & Infectious Diseases Aims and scope Submit manuscript

Abstract

Adequate empirical antimicrobial coverage is instrumental in clinical management of community-onset Enterobacteriaceae bacteraemia in areas with high ESBL prevalence, while balancing the risk of carbapenem overuse and emergence of carbapenem-resistant organisms. It is unknown whether machine learning offers additional advantages to conventional statistical methods in prediction of ESBL production. To develop a validated model to predict ESBL production in Enterobacteriaceae causing community-onset bacteraemia. 5625 patients with community-onset bacteraemia caused by Escherichia coli, Klebsiella species and Proteus mirabilis during 1 January 2015–31 December 2019 from three regional hospitals in Hong Kong were included in the analysis, after exclusion of blood cultures obtained beyond 48 h of admission. The prevalence of ESBL-producing Enterobacteriaceae was 23.7% (1335/5625). Deep neural network and other machine learning algorithms were compared against conventional statistical model via multivariable logistic regression. Primary outcomes compared consisted of predictive model area under curve of receiver-operator characteristic curve (AUC), and macro-averaged F1 score. Secondary outcomes included sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Deep neural network yielded an AUC of 0.761 (95% CI 0.725–0.797) and F1 score of 0.661 (95% CI 0.633–0.689), which was superior to logistic regression (AUC 0.667 (95% CI 0.627–0.707), F1 score 0.596 (95% CI 0.567–0.625)). Deep neural network had a specificity of 91.5%, sensitivity of 37.5%, NPV of 82.5%, and PPV of 57.9%. Deep neural network is superior to logistic regression in predicting ESBL production in Enterobacteriaceae causing community-onset bacteraemia in high-ESBL prevalence area. Machine learning offers clinical utility in guiding judicious empirical antibiotics use.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

Data availability

The datasets generated during and/or analysed during the current study are not publicly available due to ethics approval restriction, but are available from the corresponding author on reasonable request.

References

  1. Melzer M, Petersen I (2007) Mortality following bacteraemic infection caused by extended spectrum beta-lactamase (ESBL) producing E. coli compared to non-ESBL producing E. coli. J Inf Secur 55(3):254–259

    Google Scholar 

  2. Harris PNA et al (2018) Effect of piperacillin-tazobactam vs meropenem on 30-day mortality for patients with E coli or Klebsiella pneumoniae bloodstream infection and ceftriaxone resistance: a randomized clinical trial. JAMA 320(10):984–994

    Article  CAS  Google Scholar 

  3. Schwaber MJ, Carmeli Y (2007) Mortality and delay in effective therapy associated with extended-spectrum beta-lactamase production in Enterobacteriaceae bacteraemia: a systematic review and meta-analysis. J Antimicrob Chemother 60(5):913–920

    Article  CAS  Google Scholar 

  4. Alqasim A, Abu Jaffal A, Alyousef AA (2018) Prevalence of multidrug resistance and extended-spectrum beta-lactamase carriage of clinical uropathogenic Escherichia coli isolates in Riyadh, Saudi Arabia. Int J Microbiol 2018:3026851

    Article  Google Scholar 

  5. Enoch DA et al (2012) Epidemiology of extended-spectrum beta-lactamase-producing Enterobacteriaceae in a UK district hospital; an observational study. J Hosp Infect 81(4):270–277

    Article  CAS  Google Scholar 

  6. Malande OO et al (2019) A ten-year review of ESBL and non-ESBL Escherichia coli bloodstream infections among children at a tertiary referral hospital in South Africa. PLoS One 14(9):e0222675

    Article  CAS  Google Scholar 

  7. Xiao T et al (2019) A retrospective analysis of risk factors and outcomes in patients with extended-spectrum beta-lactamase-producing Escherichia coli bloodstream infections. J Glob Antimicrob Resist 17:147–156

    Article  Google Scholar 

  8. Abrar S et al (2018) Prevalence of extended-spectrum-beta-lactamase-producing Enterobacteriaceae: first systematic meta-analysis report from Pakistan. Antimicrob Resist Infect Control 7:26

    Article  Google Scholar 

  9. Goodman KE et al (2016) A clinical decision tree to predict whether a bacteremic patient is infected with an extended-spectrum beta-lactamase-producing organism. Clin Infect Dis 63(7):896–903

    Article  CAS  Google Scholar 

  10. Augustine MR et al (2017) Clinical risk score for prediction of extended-spectrum beta-lactamase-producing Enterobacteriaceae in bloodstream isolates. Infect Control Hosp Epidemiol 38(3):266–272

    Article  Google Scholar 

  11. Holmgren A, Ljung A, Bremell D (2020) An easy-to-use scoring system for predicting bacteraemia with third-generation cephalosporin-resistant Enterobacterales in a low-resistance setting. Infect Dis (Lond) 52(4):242–248

    Article  Google Scholar 

  12. Lee CH et al (2017) A simple scoring algorithm predicting extended-spectrum beta-lactamase producers in adults with community-onset monomicrobial Enterobacteriaceae bacteremia: matters of frequent emergency department users. Medicine (Baltimore) 96(16):e6648

    Article  CAS  Google Scholar 

  13. Kim M et al (2019) Clinical prediction score for community-onset bloodstream infections caused by extended-spectrum beta-lactamase-producing Escherichia coli and Klebsiella species. J Korean Med Sci 34(14):e116

    Article  Google Scholar 

  14. Food and Health Bureau, T.G.o.t.H.K.S.A.R. (2016) Chapter 1 - Hong Kong Healthcare System and Healthcare Professionals. Food and Health Bureau of the Government of HKSAR. https://www.fhb.gov.hk/download/press_and_publications/otherinfo/180500_sr/e_ch1.pdf. Accessed 27 Apr 2020

  15. Lundberg SM, Lee S-I (2017) A Unified Approach to Interpreting Model Predictions. In: 31st Conference on Neural Information Processing Systems (NIPS 2017). Long Beach, CA, USA

  16. DeLong ER, DeLong DM, Clarke-Pearson DL (1988) Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44(3):837–845

    Article  CAS  Google Scholar 

  17. Razazi K et al (2012) Clinical impact and risk factors for colonization with extended-spectrum beta-lactamase-producing bacteria in the intensive care unit. Intensive Care Med 38(11):1769–1778

    Article  Google Scholar 

  18. Tian X et al (2018) Epidemiology of and risk factors for infection with extended-spectrum beta-lactamase-producing carbapenem-resistant Enterobacteriaceae: results of a double case-control study. Infect Drug Resist 11:1339–1346

    Article  CAS  Google Scholar 

  19. Ben-Ami R et al (2009) A multinational survey of risk factors for infection with extended-spectrum beta-lactamase-producing enterobacteriaceae in nonhospitalized patients. Clin Infect Dis 49(5):682–690

    Article  Google Scholar 

  20. Rahimian J et al (2004) Pyogenic liver abscess: recent trends in etiology and mortality. Clin Infect Dis 39(11):1654–1659

    Article  Google Scholar 

  21. Siu LK et al (2012) Klebsiella pneumoniae liver abscess: a new invasive syndrome. Lancet Infect Dis 12(11):881–887

    Article  Google Scholar 

  22. Silver D et al (2016) Mastering the game of go with deep neural networks and tree search. Nature 529(7587):484–489

    Article  CAS  Google Scholar 

  23. Brown TB et al (2020) Language models are few-shot learners. arXiv. 2005.14165

  24. Rybak MJ et al (2020) Therapeutic monitoring of vancomycin for serious methicillin-resistant Staphylococcus aureus infections: a revised consensus guideline and review by the American Society of Health-System Pharmacists, the Infectious Diseases Society of America, the Pediatric Infectious Diseases Society, and the Society of Infectious Diseases Pharmacists. Am J Health Syst Pharm 77(11):835–864

    Article  Google Scholar 

  25. Cheng L et al (2017). Piperacillin-tazobactam versus other antibacterial agents for treatment of bloodstream infections due to AmpC β-lactamase-producing Enterobacteriaceae. Antimicrob Agents Chemother 61(6):e00276

  26. Tan SH et al (2020) Outcomes of treating AmpC-producing Enterobacterales bacteraemia with carbapenems vs. non-carbapenems. Int J Antimicrob Agents 55(2):105860

Download references

Author information

Authors and Affiliations

  1. Department of Microbiology, Prince of Wales Hospital, Shatin, Hong Kong

    Alfred Lok Hang Lee, Chun Wai Wong, Viola Chi Ying Chow & Raymond Wai Man Lai

  2. Department of Anatomical and Cellular Pathology, Chinese University of Hong Kong, Shatin, Hong Kong

    Curtis Chun Kit To, Angus Lang Sun Lee & Ronald Cheong Kin Chan

  3. Department of Orthopaedics and Traumatology, University of Hong Kong, Pok Fu Lam, Hong Kong

    Janus Siu Him Wong

Authors
  1. Alfred Lok Hang Lee
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Curtis Chun Kit To
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Angus Lang Sun Lee
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Ronald Cheong Kin Chan
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Janus Siu Him Wong
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Chun Wai Wong
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Viola Chi Ying Chow
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Raymond Wai Man Lai
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Alfred Lee: conceptualisation, methodology, data curation, formal analysis, writing–original draft preparation. Curtis To: software, validation, formal analysis. Angus Lee: software, validation, formal analysis. Ronald Chan: methodology, formal analysis. Janus Wong: methodology, data curation, writing–original draft preparation. Chun Wai Wong: data curation. Viola Chow: writing–review and editing, supervision. Raymond Lai: writing–review and editing, supervision.

Corresponding author

Correspondence to Alfred Lok Hang Lee.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

ESM 1

(DOCX 3284 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Lee, A.L.H., To, C.C.K., Lee, A.L.S. et al. Deep learning model for prediction of extended-spectrum beta-lactamase (ESBL) production in community-onset Enterobacteriaceae bacteraemia from a high ESBL prevalence multi-centre cohort. Eur J Clin Microbiol Infect Dis 40, 1049–1061 (2021). https://doi.org/10.1007/s10096-020-04120-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10096-020-04120-2

Search

Navigation