Abstract
Purpose
We aimed to develop a machine-learning (ML) algorithm that can predict intensive care unit (ICU)-acquired bloodstream infections (BSI) among patients suspected of infection in the ICU.
Methods
The study was based on patients’ electronic health records at Beth Israel Deaconess Medical Center (BIDMC) in Boston, Massachusetts, USA, and at Rambam Health Care Campus (RHCC), Haifa, Israel. We included adults from whom blood cultures were collected for suspected BSI at least 48 h after admission. Clinical data, including time-series variables and their interactions, were analyzed by an ML algorithm at each site. Prediction ability for ICU-acquired BSI was assessed by the area under the receiver operating characteristics (AUROC) of ten-fold cross-validation and validation sets with 95% confidence intervals.
Results
The datasets comprised 2351 patients from BIDMC (151 with BSI) and 1021 from RHCC (162 with BSI). The median (inter-quartile range) age was 62 (51–75) and 56 (38–69) years, respectively; the median Acute Physiology and Chronic Health Evaluation II scores were 26 (21–32) and 24 (20–29), respectively. The means of the cross-validation AUROCs were 0.87 ± 0.02 for BIDMC and 0.93 ± 0.03 for RHCC. AUROCs of 0.89 ± 0.01 and 0.92 ± 0.02 were maintained in both centers with internal validation, while external validation deteriorated. Valuable predictors were mainly the trends of time-series variables such as laboratory results and vital signs.
Conclusion
An ML approach that uses temporal and site-specific data achieved high performance in recognizing BC samples with a high probability for ICU-acquired BSI.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Vincent J, Sakr Y, Sprung C et al (2006) Sepsis Occurence in Acutely Ill Patients Investigators. Sepsis in European intensive care units: results of the SOAP study. Crit Care Med 34:344–353. https://doi.org/10.1097/01.CCM.0000194725.48928.3A
Wisplinghoff H, Bischoff T, Tallent SM et al (2004) Nosocomial bloodstream infections in us hospitals: analysis of 24,179 cases from a prospective nationwide surveillance study. Clin Infect Dis 39:309–317. https://doi.org/10.1086/421946
Savage RD, Fowler RA, Rishu AH et al (2016) The effect of inadequate initial empiric antimicrobial treatment on mortality in critically ill patients with bloodstream infections: a multi-centre retrospective cohort study. PLoS One 11:e0154944. https://doi.org/10.1371/journal.pone.0154944
Ibrahim EH, Sherman G, Ward S et al (2000) The influence of inadequate antimicrobial treatment of bloodstream infections on patient outcomes in the ICU setting. Chest 118:146–155. https://doi.org/10.1378/chest.118.1.146
Bartlett JG (2004) Adequacy of early empiric antibiotic treatment and survival in severe sepsis: experience from the MONARCS trial. Clin Infect Dis 38:284–288. https://doi.org/10.1086/379825
Shukrallah B, Hanna H, Hachem R et al (2007) Correlation between early clinical response after catheter removal and diagnosis of catheter-related bloodstream infection. Diagn Microbiol Infect Dis 58:453–457. https://doi.org/10.1016/j.diagmicrobio.2007.03.012
Adrie C, Garrouste-Orgeas M, Ibn Essaied W et al (2017) Attributable mortality of ICU-acquired bloodstream infections: impact of the source, causative micro-organism, resistance profile and antimicrobial therapy. J Infect 74:131–141. https://doi.org/10.1016/j.jinf.2016.11.001
Brooks D, Smith A, Young D et al (2016) Mortality in intensive care: the impact of bacteremia and the utility of systemic inflammatory response syndrome. Am J Infect Control 44:1291–1295. https://doi.org/10.1016/j.ajic.2016.04.214
Eliakim-Raz N, Bates DW, Leibovici L (2015) Predicting bacteraemia in validated models-a systematic review. Clin Microbiol Infect 21:295–301. https://doi.org/10.1016/j.cmi.2015.01.023
Jensen JU, Hein L, Lundgren B et al (2011) Procalcitonin-guided interventions against infections to increase early appropriate antibiotics and improve survival in the intensive care unit: a randomized trial. Crit Care Med 39:2048–2058. https://doi.org/10.1097/CCM.0b013e31821e8791
de Jong E, van Oers JA, Beishuizen A et al (2016) Efficacy and safety of procalcitonin guidance in reducing the duration of antibiotic treatment in critically ill patients: a randomised, controlled, open-label trial. Lancet Infect Dis 16:819–827. https://doi.org/10.1016/S1473-3099(16)00053-0
Rajkomar A, Dean J, Kohane I (2019) Machine learning in medicine. N Engl J Med 380:1347–1358. https://doi.org/10.1056/NEJMra1814259
Bailly S, Meyfroidt G, Timsit J-F (2018) What’s new in ICU in 2050: big data and machine learning. Intensive Care Med 44:1524–1527. https://doi.org/10.1007/s00134-017-5034-3
Komorowski M (2019) Artificial intelligence in intensive care: are we there yet? Intensive Care Med. https://doi.org/10.1007/s00134-019-05662-6
Johnson AEW, Pollard TJ, Shen L et al (2016) MIMIC-III, a freely accessible critical care database. Sci Data. https://doi.org/10.1038/sdata.2016.35
Almeida LB (2018) The fractional Fourier transform and time-frequency representations. IEEE J Mag https://ieeexplore.ieee.org/abstract/document/330368
Saeys Y, Inza I, Larranaga P (2007) A review of feature selection techniques in bioinformatics. Bioinformatics 23:2507–2517. https://doi.org/10.1093/bioinformatics/btm344
Polikar R (2006) Ensemble based systems in decision making. IEEE Circuits Syst Mag 6:21–45. https://doi.org/10.1109/MCAS.2006.1688199
LeDell E, Petersen M, van der Laan M (2015) Computationally efficient confidence intervals for cross-validated area under the ROC curve estimates. Electron J Stat 9:1583–1607. https://doi.org/10.1214/15-EJS1035
Paul M, Andreassen S, Nielsen AD et al (2006) Prediction of bacteremia using TREAT, a computerized decision-support system. Clin Infect Dis 42:1274–1282. https://doi.org/10.1086/503034
Ratzinger F, Haslacher H, Perkmann T et al (2018) Machine learning for fast identification of bacteraemia in SIRS patients treated on standard care wards: a cohort study. Sci Rep 8:1–8. https://doi.org/10.1038/s41598-018-30236-9
Beam AL, Kohane IS (2018) Big data and machine learning in health care. JAMA 319:1317. https://doi.org/10.1001/jama.2017.18391
Rajkomar A, Oren E, Chen K et al (2018) Scalable and accurate deep learning for electronic health records. Digit Med 1:18. https://doi.org/10.1038/s41746-018-0029-1
Vincent J-LL, Rello J, Marshall J et al (2009) International study of the prevalence and outcomes of infection in intensive care units. JAMA 302:2323–2329. https://doi.org/10.1001/jama.2009.1754
Kariv G, Paul M, Shani V et al (2013) Benchmarking inappropriate empirical antibiotic treatment. Clin Microbiol Infect 19:629–633. https://doi.org/10.1111/j.1469-0691.2012.03965.x
Tabah A, Koulenti D, Laupland K et al (2012) Characteristics and determinants of outcome of hospital-acquired bloodstream infections in intensive care units: the EUROBACT International Cohort Study. Intensive Care Med 38:1930–1945. https://doi.org/10.1007/s00134-012-2695-9
Lee A, Mirrett S, Reller LB, Weinstein MP (2007) Detection of bloodstream infections in adults: how many blood cultures are needed? J Clin Microbiol 45:3546–3548. https://doi.org/10.1128/JCM.01555-07
Cockerill FR, Wilson JW, Vetter EA et al (2004) Optimal testing parameters for blood cultures. Clin Infect Dis 38:1724–1730. https://doi.org/10.1086/421087
Cheng MP, Stenstrom R, Paquette K et al (2019) Blood culture results before and after antimicrobial administration in patients with severe manifestations of sepsis: a diagnostic study. Ann Intern Med. https://doi.org/10.7326/M19-1696
Georgevici AI, Terblanche M (2019) Neural networks and deep learning: a brief introduction. Intensive Care Med 45:712–714. https://doi.org/10.1007/s00134-019-05537-w
Rhodes A, Evans LE, Alhazzani W et al (2017) Surviving sepsis campaign: international guidelines for management of sepsis and septic shock: 2016. Crit Care Med 45:486–552. https://doi.org/10.1097/CCM.0000000000002255
Levy MM, Evans LE, Rhodes A (2018) The Surviving Sepsis Campaign Bundle: 2018 update. Intensive Care Med 44:925–928. https://doi.org/10.1007/s00134-018-5085-0
Paul M, Bishara J, Yahav D, et al (2015) Trimethoprim-sulfamethoxazole versus vancomycin for severe infections caused by meticillin resistant Staphylococcus aureus: randomised controlled trial. BMJ https://www.ncbi.nlm.nih.gov/pubmed/25977146
Vidal L, Gafter-Gvili A, Borok S et al (2007) Efficacy and safety of aminoglycoside monotherapy: systematic review and meta-analysis of randomized controlled trials. J Antimicrob Chemother. https://doi.org/10.1093/jac/dkm193
Acknowledgements
We thank Belina Neuberger for editing the manuscript.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interest
On behalf of all authors, the corresponding author states that there is no conflict of interest, nor did they receive any financial support for the current study. The code of the models was uploaded into GitHub under “ICU-acquired BSI prediction model”.
Ethical approval
The study was approved by the independent ethics committee of Rambam Health Care Campus.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Roimi, M., Neuberger, A., Shrot, A. et al. Early diagnosis of bloodstream infections in the intensive care unit using machine-learning algorithms . Intensive Care Med 46, 454–462 (2020). https://doi.org/10.1007/s00134-019-05876-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00134-019-05876-8