Abstract:Background. Patients with Gram-negative bloodstream infections are at risk of serious adverse outcomes without active treatment, but identifying who has antimicrobial resistance (AMR) to target empirical treatment is challenging. Methods. We used XGBoost machine learning models to predict the presence of antimicrobial resistance to seven antibiotics in patients with Enterobacterales bloodstream infection. Models were trained using hospital and community data available at the time blood cultures were obtained from Oxfordshire, UK, between 01-January-2017 and 31-December-2021. Model performance was compared to final microbiology results using test datasets from 01-January-2022 to 31-December-2023 and with clinicians' prescribing. Findings. 4709 infection episodes were used for model training and evaluation; antibiotic resistance rates ranged from 7-67%. In held-out test data, resistance prediction performance was similar for the seven antibiotics (AUCs 0.680 [95%CI 0.641-0.720] to 0.737 [0.674-0.797]). Performance improved for most antibiotics when species data were included as model inputs (AUCs 0.723 [0.652-0.791] to 0.827 [0.797-0.857]). In patients treated with a beta-lactam, clinician prescribing led to 70% receiving an active beta-lactam: 44% were over-treated (broader spectrum treatment than needed), 26% optimally treated (narrowest spectrum active agent), and 30% under-treated (inactive beta-lactam). Model predictions without species data could have led to 79% of patients receiving an active beta-lactam: 45% over-treated, 34% optimally treated, and 21% under-treated. Interpretation. Predicting AMR in bloodstream infections is challenging for both clinicians and models. Despite modest performance, machine learning models could still increase the proportion of patients receiving active empirical treatment by up to 9% over current clinical practice in an environment prioritising antimicrobial stewardship. Funding. National Institute of Health Research (NIHR) Oxford Biomedical Research Centre, NIHR Health Protection Research Unit in Healthcare-associated Infection and Antimicrobial Resistance.

What problem does this paper attempt to address?

The paper attempts to address the problem of predicting antimicrobial resistance (AMR) in Enterobacteriaceae bloodstream infections to guide empirical treatment. Specifically, the researchers utilized a machine learning model (XGBoost) to predict resistance to seven commonly used antibiotics, thereby helping doctors choose the appropriate antibiotics more accurately before obtaining the final microbiological results. ### Background Bloodstream infections caused by Gram-negative bacteria can lead to severe adverse outcomes, and timely and effective antimicrobial treatment can save lives and improve patient prognosis. However, determining which patients have antimicrobial resistance to target empirical treatment is very challenging. Traditional antimicrobial treatment guidelines, while aiming to maximize effective treatment, may also lead to the overuse of broad-spectrum antibiotics, thereby driving the development of antimicrobial resistance. ### Methods The researchers used the XGBoost machine learning model, trained on hospital and community data collected from Oxfordshire, UK, between 2017 and 2021, to predict resistance to seven antibiotics. The model's performance was evaluated using test datasets from 2022 and 2023 and compared with actual prescriptions by clinicians. ### Results - **Model Performance**: In the 2022 test dataset, the model's predictive performance for the seven antibiotics was similar, with AUC values ranging from 0.680 to 0.737. When species information was included as model input, performance improved, with AUC values reaching up to 0.827. - **Clinical Comparison**: Among patients treated with β-lactam antibiotics, clinicians' prescriptions resulted in 70% of patients receiving effective β-lactam antibiotic treatment, with 44% overtreated, 26% optimally treated, and 30% undertreated. The model's predictions could increase the proportion of patients receiving effective treatment to 79%, with 45% overtreated, 34% optimally treated, and 21% undertreated. ### Interpretation Although the performance of the machine learning model is relatively limited, it can still increase the proportion of patients receiving effective empirical treatment in current clinical practice by up to 9%. This is particularly important in environments prioritizing antimicrobial stewardship. ### Significance This study demonstrates the potential of machine learning models in predicting antimicrobial resistance in bloodstream infections. Despite limited performance, they can still significantly improve patient treatment outcomes, providing important references for future clinical applications.

Machine learning to predict antibiotic susceptibility in Enterobacterales bloodstream infections compared to clinician prescribing

Predicting future hospital antimicrobial resistance prevalence using machine learning

Integrating whole genome sequencing and machine learning for predicting antimicrobial resistance in critical pathogens: a systematic review of antimicrobial susceptibility tests

Machine learning model for the prediction of gram-positive and gram-negative bacterial bloodstream infection based on routine laboratory parameters

Prediction of Antibiotic Susceptibility in E. coli Isolates Using Machine Learning

Evaluation of parameters affecting performance and reliability of machine learning-based antibiotic susceptibility testing from whole genome sequencing data

Machine learning-based antibiotic resistance prediction models: An updated systematic review and meta-analysis

Machine learning and feature extraction for rapid antimicrobial resistance prediction of Acinetobacter baumannii from whole-genome sequencing data

Identification of key drivers of antimicrobial resistance in Enterococcus using machine learning

Prediction of ciprofloxacin resistance in hospitalized patients using machine learning

Machine Learning Techniques for Antimicrobial Resistance Prediction of Pseudomonas Aeruginosa from Whole Genome Sequence Data

Prediction of antimicrobial resistance based on whole-genome sequencing and machine learning

Proof of concept study on early forecasting of antimicrobial resistance in hospitalized patients using machine learning and simple bacterial ecology data

Applying Machine Learning for Antibiotic Development and Prediction of Microbial Resistance

Personalized antibiograms for machine learning driven antibiotic selection

Using interpretable machine learning to predict bloodstream infection and antimicrobial resistance in patients admitted to ICU: Early alert predictors based on EHR data to guide antimicrobial stewardship

Prediction of antibiotic resistance in Escherichia coli from large-scale pan-genome data

Battling Antibiotic Resistance: Can Machine Learning Improve Prescribing?

Moving from Predicting Hospital Deaths by Antibiotic-Resistant Bloodstream Bacteremia Toward Actionable Risk Reduction Using Machine Learning on Electronic Health Records.

Predicting Antimicrobial Resistance in the Intensive Care Unit

Predicting Beta-Lactam Target Non-Attainment in ICU Patients at Treatment Initiation: Development and External Validation of Three Novel (Machine Learning) Models