Search

The aim of the study was to build a machine learning-based predictive model to discriminate between hospitalized patients at low risk and high risk of bloodstream infection (BSI). A Data Mart including all patients hospitalized between January 2016 and December 2019 with suspected BSI was built. Multivariate logistic regression was applied to develop a clinically interpretable machine learning predictive model. The model was trained on 2016–2018 data and tested on 2019 data. A feature selection based on a univariate logistic regression first selected candidate predictors of BSI. A multivariate logistic regression with stepwise feature selection in five-fold cross-validation was applied to express the risk of BSI. A total of 5660 hospitalizations (4026 and 1634 in the training and the validation subsets, respectively) were included. Eleven predictors of BSI were identified. The performance of the model in terms of AUROC was 0.74. Based on the interquartile predicted risk score, 508 (31.1%) patients were defined as being at low risk, 776 (47.5%) at medium risk, and 350 (21.4%) at high risk of BSI. Of them, 14.2% (72/508), 30.8% (239/776), and 64% (224/350) had a BSI, respectively. The performance of the predictive model of BSI is promising. Computational infrastructure and machine learning models can help clinicians identify people at low risk for BSI, ultimately supporting an antibiotic stewardship approach. [ABSTRACT FROM AUTHOR]

Objectives To assess the impact of piperacillin/tazobactam MICs on in-hospital 30 day mortality in patients with third-generation cephalosporin-resistant Escherichia coli bloodstream infection treated with piperacillin/tazobactam, compared with those treated with carbapenems. Methods A multicentre retrospective cohort study was conducted in three large academic hospitals in Italy between 2018 and 2022. The study population comprised patients with monomicrobial third-generation cephalosporin-resistant E. coli bloodstream infection, who received either piperacillin/tazobactam or carbapenem therapy within 48 h of blood culture collection. The primary outcome was in-hospital 30 day all-cause mortality. A propensity score was used to estimate the likelihood of receiving empirical piperacillin/tazobactam treatment. Cox regression models were performed to ascertain risk factors independently associated with in-hospital 30 day mortality. Results Of the 412 consecutive patients included in the study, 51% received empirical therapy with piperacillin/tazobactam, while 49% received carbapenem therapy. In the propensity-adjusted multiple Cox model, the Pitt bacteraemia score [HR 1.38 (95% CI, 0.85–2.16)] and piperacillin/tazobactam MICs of 8 mg/L [HR 2.35 (95% CI, 1.35–3.95)] and ≥16 mg/L [HR 3.69 (95% CI, 1.86–6.91)] were significantly associated with increased in-hospital 30 day mortality, while the empirical use of piperacillin/tazobactam was not found to predict in-hospital 30 day mortality [HR 1.38 (95% CI, 0.85–2.16)]. Conclusions Piperacillin/tazobactam use might not be associated with increased mortality in treating third-generation cephalosporin-resistant E. coli bloodstream infections when the MIC is <8 mg/L. [ABSTRACT FROM AUTHOR]