Search

Aims: Hypokalemia is a common complication following traumatic brain injury, which may complicate treatment and lead to unfavorable outcomes. Identifying patients at risk of hypokalemia on the first day of admission helps to implement prophylactic treatment, reduce complications, and improve prognosis. Methods: This multicenter retrospective study was performed between January 2017 and December 2020 using the electronic medical records of patients admitted due to traumatic brain injury. A propensity score matching approach was adopted with a ratio of 1:1 to overcome overfitting and data imbalance during subgroup analyses. Five machine learning algorithms were applied to generate a best‐performed prediction model for in‐hospital hypokalemia. The internal fivefold cross‐validation and external validation were performed to demonstrate the interpretability and generalizability. Results: A total of 4445 TBI patients were recruited for analysis and model generation. Hypokalemia occurred in 46.55% of recruited patients and the incidences of mild, moderate, and severe hypokalemia were 32.06%, 12.69%, and 1.80%, respectively. Hypokalemia was associated with increased mortality, while severe hypokalemia cast greater impacts. The logistic regression algorithm had the best performance in predicting decreased serum potassium and moderate‐to‐severe hypokalemia, with an AUC of 0.73 ± 0.011 and 0.74 ± 0.019, respectively. The prediction model was further verified using two external datasets, including our previous published data and the open‐assessed Medical Information Mart for Intensive Care database. Linearized calibration curves showed no statistical difference (p > 0.05) with perfect predictions. Conclusions: The occurrence of hypokalemia following traumatic brain injury can be predicted by first hospitalization day records and machine learning algorithms. The logistic regression algorithm showed an optimal predicting performance verified by both internal and external validation. [ABSTRACT FROM AUTHOR]

Abstract Background Medical personnel in hospitals often works under great physical and mental strain. In medical decision-making, errors can never be completely ruled out. Several studies have shown that between 50 and 60% of adverse events could have been avoided through better organization, more attention or more effective security procedures. Critical situations especially arise during interdisciplinary collaboration and the use of complex medical technology, for example during surgical interventions and in perioperative settings (the period of time before, during and after surgical intervention). Methods In this paper, we present an ontology and an ontology-based software system, which can identify risks across medical processes and supports the avoidance of errors in particular in the perioperative setting. We developed a practicable definition of the risk notion, which is easily understandable by the medical staff and is usable for the software tools. Based on this definition, we developed a Risk Identification Ontology (RIO) and used it for the specification and the identification of perioperative risks. Results An agent system was developed, which gathers risk-relevant data during the whole perioperative treatment process from various sources and provides it for risk identification and analysis in a centralized fashion. The results of such an analysis are provided to the medical personnel in form of context-sensitive hints and alerts. For the identification of the ontologically specified risks, we developed an ontology-based software module, called Ontology-based Risk Detector (OntoRiDe). Conclusions About 20 risks relating to cochlear implantation (CI) have already been implemented. Comprehensive testing has indicated the correctness of the data acquisition, risk identification and analysis components, as well as the web-based visualization of results.

Background: Medical personnel in hospitals often works under great physical and mental strain. In medical decision-making, errors can never be completely ruled out. Several studies have shown that between 50 and 60% of adverse events could have been avoided through better organization, more attention or more effective security procedures. Critical situations especially arise during interdisciplinary collaboration and the use of complex medical technology, for example during surgical interventions and in perioperative settings (the period of time before, during and after surgical intervention). Methods: In this paper, we present an ontology and an ontology-based software system, which can identify risks across medical processes and supports the avoidance of errors in particular in the perioperative setting. We developed a practicable definition of the risk notion, which is easily understandable by the medical staff and is usable for the software tools. Based on this definition, we developed a Risk Identification Ontology (RIO) and used it for the specification and the identification of perioperative risks. Results: An agent system was developed, which gathers risk-relevant data during the whole perioperative treatment process from various sources and provides it for risk identification and analysis in a centralized fashion. The results of such an analysis are provided to the medical personnel in form of context-sensitive hints and alerts. For the identification of the ontologically specified risks, we developed an ontology-based software module, called Ontology-based Risk Detector (OntoRiDe). Conclusions: About 20 risks relating to cochlear implantation (CI) have already been implemented. Comprehensive testing has indicated the correctness of the data acquisition, risk identification and analysis components, as well as the web-based visualization of results. [ABSTRACT FROM AUTHOR]

Aims Beta-blockers vary in pharmacodynamics and pharmacokinetic properties. It is unknown whether specific types are associated with increased perioperative risks. We evaluated perioperative risks associated with beta-blocker subtypes, overall and in patient subgroups. Methods and results We performed a Danish Nationwide cohort study, 2005-2011, of patients treated chronically with beta blocker (atenolol, bisoprolol, carvedilol, metoprolol, propranolol, or other) prior to non-cardiac surgery. Risks of 30-day all-cause mortality (ACM) and 30-day major adverse cardiovascular events (MACE) were estimated using adjusted logistic regression models and odds ratios with 95% confidence intervals. We identified 61 660 patients, most frequently treated with metoprolol (67% of patients, mean age 69 years, 49% males), atenolol (10% of patients, mean age 68 years, 36% males), or carvedilol (9% of patients, mean age 68 years, 60% males). The crude incidences of ACM and MACE were 4.1 and 3.5% in patients with metoprolol, 3.0 and 2.3% with atenolol, and 4.8 and 4.6% with carvedilol. In adjusted models, risks were not significantly different with atenolol (ACM; 1.10 [0.92-1.32], MACE; 1.08 [0.90-1.31]) or carvedilol (ACM; 0.99 [0.85-1.16], MACE; 1.07 [0.92-1.25]), compared with metoprolol. Risks of ACM were significantly lower in prior myocardial infarction patients treated with carvedilol (0.62 [0.43-0.87]) and no different in patients with uncomplicated hypertension (1.41 [0.83-2.40]). Risks did not differ in analyses stratified by age, surgery priority, duration of anaesthesia or surgery risk (all P for interaction >0.05). Conclusion Risks of ACM and MACE did not systematically differ by beta-blocker subtype. Findings may guide clinical practice and future trials. [ABSTRACT FROM AUTHOR]