The following is a summary of “A machine learning-based risk score for prediction of infective endocarditis among patients with Staphylococcus aureus bacteraemia – The SABIER score,” published in February 2024, issue of Infectious Disease by Lai et al.
Researchers conducted a retrospective study assessing early Staphylococcus aureus infective endocarditis (SA-IE) risk among Staphylococcus aureus bacteraemia (SAB) patients to guide clinical management and develop novel risk scores independent of clinical judgment.
They used hospitalized SAB patient data (2009 to 2019), selecting predictive variables using a random forest risk scoring model. The dataset was divided for derivation and validation, evaluating areas under the receiver operating characteristic curve (AUCROC).
The results showed that of 15,741 SAB patients, 4.18% had SA-IE. AUCROC was 0.74 (95% CI 0.70-0.76), with a negative predictive value of 0.980 (95% CI 0.977-0.983). The four most discriminative features were age, infective endocarditis history, valvular heart disease, and community-onset.
The results showed 15,741 patients diagnosed with SAB, with 4.18% exhibiting SA-IE. The AUCROC stood at 0.74 (95% CI 0.70-0.76), with a negative predictive value of 0.980 (95% CI 0.977–0.983). Age, history of IE valvular heart disease, and community-onset status emerged as the top four distinguishing factors.
Investigators concluded that the novel risk score developed exhibited strong performance. It offers an objective tool for early SAB assessment, surpassing reliance on subjective clinical judgment.
Source: academic.oup.com/jid/advance-article-abstract/doi/10.1093/infdis/jiae080/7616139
