小野 将平

BACKGROUND: Clinically important gastrointestinal bleeding (CIGIB) is a serious complication in critically ill patients, contributing to prolonged ICU stays and increased mortality. Despite efforts to identify high-risk patients, no previous studies have employed machine learning models to predict CIGIB during ICU stay or identify key predictors in this context. METHODS: This single-center retrospective study included ICU patients aged 18 years or older admitted between 2017 and 2024. Patients with ICU stays of less than 24 hours or GIB within 24 hours of admission were excluded. Machine learning models, including XGBoost, Random Forest, and L1-regularized logistic regression, were trained using patient data from the first 24 hours of ICU admission. Model performance was assessed using AUROC, precision, recall, and F1 scores. Shapley Additive Explanations (SHAP) were employed to evaluate key predictors. RESULTS: A total of 7,357 ICU patients were included, of whom 171 (2.3%) experienced CIGIB. The XGBoost model demonstrated the highest predictive performance with an AUROC of 0.84. Key predictors included APACHE III scores, hematocrit levels, APTT, creatinine and respiratory rate, while invasive mechanical ventilation and stress ulcer prophylaxis within the first 24 hours of ICU admission did not rank among the top 20 predictors based on SHAP values. CONCLUSIONS: This study represents the first application of machine learning for predicting CIGIB in ICU patients, providing valuable insights into risk stratification. The model demonstrated high predictive accuracy and interpretability, highlighting its potential to guide early intervention and prophylaxis. Further multi-center studies and interventional trials are needed to validate these findings and refine clinical risk prediction strategies.

Background Previous studies have demonstrated a correlation between management by intensivists and a decrease in hospital stay and mortality, yet the underlying reason remains unknown. Using open data from the National Database of Health Insurance Claims and Specific Health Checkups of Japan (NDB) and other databses, the present study aimed to explore the relationship between inotrope and vasoconstrictor use and the number of intensivists. Materials and methods Cardiovascular agonists listed in the 2020 NDB for which the total dose was known were included for analysis. Trends in cardiovascular agonist use over six years were then graphically assessed, and a linear regression model with the use of each target drug per prefecture as the objective variable in the 2020 data was created to analyze the impact of intensivists on drug use. Results A total of 61 drugs were classified into eight groups based on their composition, and drug use in each of the 47 prefectures was tabulated. Both the rate of use and cost showed a yearly decrease for dopamine but a yearly increase for norepinephrine. Multivariable analysis indicated that the number of intensivists was only significant for dopamine, which had a coefficient of -310 (95% CI: -548 to -72, p = 0.01) but that no such trend was evident for the other drugs. Conclusions The results demonstrated that an increasing number of intensivists in each prefecture correlated with decreasing use of dopamine, possibly explaining the improved outcomes observed in closed ICUs led by intensivists. Further research is warranted to establish causality.