The effect of COVID-19 vaccination status on all-cause mortality in patients hospitalised with COVID-19 in Hungary during the delta wave of the pandemic.

The high mortality of patients with coronavirus disease 2019 (COVID-19) is effectively reduced by vaccination. However, the effect of vaccination on mortality among hospitalised patients is under-researched. Thus, we investigated the effect of a full primary or an additional booster vaccination on in-hospital mortality among patients hospitalised with COVID-19 during the delta wave of the pandemic. This retrospective cohort included all patients (n = 430) admitted with COVID-19 at Semmelweis University Department of Medicine and Oncology in 01/OCT/2021-15/DEC/2021. Logistic regression models were built with COVID-19-associated in-hospital/30 day-mortality as outcome with hierarchical entry of predictors of vaccination, vaccination status, measures of disease severity, and chronic comorbidities. Deceased COVID-19 patients were older and presented more frequently with cardiac complications, chronic kidney disease, and active malignancy, as well as higher levels of inflammatory markers, serum creatinine, and lower albumin compared to surviving patients (all p < 0.05). However, the rates of vaccination were similar (52-55%) in both groups. Based on the fully adjusted model, there was a linear decrease of mortality from no/incomplete vaccination (ref) through full primary (OR 0.69, 95% CI: 0.39-1.23) to booster vaccination (OR 0.31, 95% CI 0.13-0.72, p = 0.006). Although unadjusted mortality was similar among vaccinated and unvaccinated patients, this was explained by differences in comorbidities and disease severity. In adjusted models, a full primary and especially a booster vaccination improved survival of patients hospitalised with COVID-19 during the delta wave of the pandemic. Our findings may improve the quality of patient provider discussions at the time of admission.

EASY-APP: An artificial intelligence model and application for early and easy prediction of severity in acute pancreatitis.

BACKGROUND: Acute pancreatitis (AP) is a potentially severe or even fatal inflammation of the pancreas. Early identification of patients at high risk for developing a severe course of the disease is crucial for preventing organ failure and death. Most of the former predictive scores require many parameters or at least 24 h to predict the severity; therefore, the early therapeutic window is often missed. METHODS: The early achievable severity index (EASY) is a multicentre, multinational, prospective and observational study (ISRCTN10525246). The predictions were made using machine learning models. We used the scikit-learn, xgboost and catboost Python packages for modelling. We evaluated our models using fourfold cross-validation, and the receiver operating characteristic (ROC) curve, the area under the ROC curve (AUC), and accuracy metrics were calculated on the union of the test sets of the cross-validation. The most critical factors and their contribution to the prediction were identified using a modern tool of explainable artificial intelligence called SHapley Additive exPlanations (SHAP). RESULTS: The prediction model was based on an international cohort of 1184 patients and a validation cohort of 3543 patients. The best performing model was an XGBoost classifier with an average AUC score of 0.81 ± 0.033 and an accuracy of 89.1%, and the model improved with experience. The six most influential features were the respiratory rate, body temperature, abdominal muscular reflex, gender, age and glucose level. Using the XGBoost machine learning algorithm for prediction, the SHAP values for the explanation and the bootstrapping method to estimate confidence, we developed a free and easy-to-use web application in the Streamlit Python-based framework (http://easy-app.org/). CONCLUSIONS: The EASY prediction score is a practical tool for identifying patients at high risk for severe AP within hours of hospital admission. The web application is available for clinicians and contributes to the improvement of the model.

Early prediction of acute necrotizing pancreatitis by artificial intelligence: a prospective cohort-analysis of 2387 cases.

Pancreatic necrosis is a consistent prognostic factor in acute pancreatitis (AP). However, the clinical scores currently in use are either too complicated or require data that are unavailable on admission or lack sufficient predictive value. We therefore aimed to develop a tool to aid in necrosis prediction. The XGBoost machine learning algorithm processed data from 2387 patients with AP. The confidence of the model was estimated by a bootstrapping method and interpreted via the 10th and the 90th percentiles of the prediction scores. Shapley Additive exPlanations (SHAP) values were calculated to quantify the contribution of each variable provided. Finally, the model was implemented as an online application using the Streamlit Python-based framework. The XGBoost classifier provided an AUC value of 0.757. Glucose, C-reactive protein, alkaline phosphatase, gender and total white blood cell count have the most impact on prediction based on the SHAP values. The relationship between the size of the training dataset and model performance shows that prediction performance can be improved. This study combines necrosis prediction and artificial intelligence. The predictive potential of this model is comparable to the current clinical scoring systems and has several advantages over them.

Comparison of plasma and urinary microRNA-483-5p for the diagnosis of adrenocortical malignancy.

INTRODUCTION: Minimally invasive circulating microRNAs might be used for the preoperative differentiation of adrenocortical carcinoma (ACC) and adrenocortical adenoma (ACA). So far, the best blood-borne microRNA biomarker of ACC is circulating hsa-miR-483-5p. The expression of urinary hsa-miR-483-5p as a non-invasive marker of malignancy and its correlation with plasma hsa-miR-483-5p, has not been investigated, yet. AIM: Our aim was to investigate the expression of urinary hsa-miR-483-5p and its correlation with its plasma counterpart. METHODS: Plasma and urinary samples from 23 ACC and 23 ACA patients were analysed using real-time RT-qPCR. To evaluate the diagnostic applicability of hsa-miR-483-5p, ROC-analysis was performed. RESULTS: Significant overexpression of hsa-miR-483-5p was observed in carcinoma patients' plasma samples compared to adenoma patients' (p < 0.0001, sensitivity: 87%, specificity: 78.3%). In urinary samples, however, no significant difference could be detected between ACC and ACA patients. CONCLUSIONS: Plasma hsa-miR-483-5p has been confirmed as significantly overexpressed in adrenocortical cancer patients and thus might be exploited as a minimally invasive preoperative marker of malignancy. The applicability of urinary hsa-miR-483-5p for the diagnosis of adrenocortical malignancy could not be confirmed.