Predictors of acute kidney injury in patients with acute decompensated heart failure in emergency departments in China

Objective This retrospective multicentre observational study was performed to assess the predictors of acute kidney injury (AKI) in patients with acute decompensated heart failure (ADHF) in emergency departments in China. Methods In total, 1743 consecutive patients with ADHF were recruited from August 2017 to January 2018. Clinical characteristics and outcomes were compared between patients with and without AKI. Predictors of AKI occurrence and underdiagnosis were assessed in multivariate regression analyses. Results Of the 1743 patients, 593 (34.0%) had AKI. AKI was partly associated with short-term all-cause mortality and cost. Cardiovascular comorbidities such as coronary heart disease, diabetes mellitus, and hypertension remained significant predictors of AKI in the univariate analysis. AKI was significantly more likely to occur in patients with a lower arterial pH, lower albumin concentration, higher creatinine concentration, and higher N-terminal pro-brain natriuretic peptide (NT-proBNP) concentration. Patients treated with inotropic agents were significantly more likely to develop AKI during their hospital stay. Conclusion This study suggests that cardiovascular comorbidities, arterial pH, the albumin concentration, the creatinine concentration, the NT-proBNP concentration, and use of inotropic agents are predictors of AKI in patients with ADHF.

Clinical Features of Kawasaki Disease during SARS-CoV-2 Epidemic: A Single-Center Retrospective Study

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) epidemic, causing coronavirus disease 2019 (COVID-19), has rapidly spread worldwide. Recently, cases of Kawasaki-like syndrome associated with COVID-19 (MIS‐C/PIMS) have been reported in the USA and Europe. However, the clinical features of Kawasaki disease (KD) in Japan during the SARS-CoV-2 epidemic remain unclear. Therefore, this retrospective study aimed to evaluate the clinical features of KD in our department during the year of the COVID-19 pandemic (2020). Data were obtained from 45 KD patients who underwent acute-phase treatment from January 2018 to December 2020. The patients were divided into three groups: (1) 2018 group, in which 18 patients receiving acute-phase treatment during 2018; (2) 2019 group, in which 17 patients receiving acute-phase treatment during 2019; and (3) 2020 group, in which 10 patients receiving acute-phase treatment during 2020. The comparison of the clinical findings, treatments, and outcomes among the three groups (2018 vs. 2019 vs. 2020 group) revealed a significant difference in the rate of the male gender (72.2% vs. 41.2% vs. 10.0%, P < 0.010), recurrence (0.0% vs. 0.0% vs. 25.0%, P = 0.015), and coronary artery lesions (16.7% vs. 0.0% vs. 0.0%, P = 0.044). No patients required ventilation assistance or inotropic agents and met the diagnostic criteria of MIS‐C/PIMS. Further studies are warranted to elucidate the clinical features of KD during the SARS-CoV-2 epidemic.

Machine Learning–Driven Models to Predict Prognostic Outcomes in Patients Hospitalized With Heart Failure Using Electronic Health Records: Retrospective Study

Background With the prevalence of cardiovascular diseases increasing worldwide, early prediction and accurate assessment of heart failure (HF) risk are crucial to meet the clinical demand. Objective Our study objective was to develop machine learning (ML) models based on real-world electronic health records to predict 1-year in-hospital mortality, use of positive inotropic agents, and 1-year all-cause readmission rate. Methods For this single-center study, we recruited patients with newly diagnosed HF hospitalized between December 2010 and August 2018 at the First Affiliated Hospital of Dalian Medical University (Liaoning Province, China). The models were constructed for a population set (90:10 split of data set into training and test sets) using 79 variables during the first hospitalization. Logistic regression, support vector machine, artificial neural network, random forest, and extreme gradient boosting models were investigated for outcome predictions. Results Of the 13,602 patients with HF enrolled in the study, 537 (3.95%) died within 1 year and 2779 patients (20.43%) had a history of use of positive inotropic agents. ML algorithms improved the performance of predictive models for 1-year in-hospital mortality (areas under the curve [AUCs] 0.92-1.00), use of positive inotropic medication (AUCs 0.85-0.96), and 1-year readmission rates (AUCs 0.63-0.96). A decision tree of mortality risk was created and stratified by single variables at levels of high-sensitivity cardiac troponin I (<0.068 μg/L), followed by percentage of lymphocytes (<14.688%) and neutrophil count (4.870×109/L). Conclusions ML techniques based on a large scale of clinical variables can improve outcome predictions for patients with HF. The mortality decision tree may contribute to guiding better clinical risk assessment and decision making.

Machine Learning–Driven Models to Predict Prognostic Outcomes in Patients Hospitalized With Heart Failure Using Electronic Health Records: Retrospective Study (Preprint)

BACKGROUND With the prevalence of cardiovascular diseases increasing worldwide, early prediction and accurate assessment of heart failure (HF) risk are crucial to meet the clinical demand. OBJECTIVE Our study objective was to develop machine learning (ML) models based on real-world electronic health records to predict 1-year in-hospital mortality, use of positive inotropic agents, and 1-year all-cause readmission rate. METHODS For this single-center study, we recruited patients with newly diagnosed HF hospitalized between December 2010 and August 2018 at the First Affiliated Hospital of Dalian Medical University (Liaoning Province, China). The models were constructed for a population set (90:10 split of data set into training and test sets) using 79 variables during the first hospitalization. Logistic regression, support vector machine, artificial neural network, random forest, and extreme gradient boosting models were investigated for outcome predictions. RESULTS Of the 13,602 patients with HF enrolled in the study, 537 (3.95%) died within 1 year and 2779 patients (20.43%) had a history of use of positive inotropic agents. ML algorithms improved the performance of predictive models for 1-year in-hospital mortality (areas under the curve [AUCs] 0.92-1.00), use of positive inotropic medication (AUCs 0.85-0.96), and 1-year readmission rates (AUCs 0.63-0.96). A decision tree of mortality risk was created and stratified by single variables at levels of high-sensitivity cardiac troponin I (<0.068 μg/L), followed by percentage of lymphocytes (<14.688%) and neutrophil count (4.870×10⁹/L). CONCLUSIONS ML techniques based on a large scale of clinical variables can improve outcome predictions for patients with HF. The mortality decision tree may contribute to guiding better clinical risk assessment and decision making.