Journal of critical care
-
Journal of critical care · Dec 2024
ReviewMachine learning for predicting mortality in adult critically ill patients with Sepsis: A systematic review.
Various Machine Learning (ML) models have been used to predict sepsis-associated mortality. We conducted a systematic review to evaluate the methodologies employed in studies to predict mortality among patients with sepsis. ⋯ ML models demonstrate a modest improvement in predicting sepsis-associated mortality. The certainty of these findings remains low due to the high risk of bias and significant heterogeneity. Studies should include comprehensive methodological details on calibration and hyperparameter selection, adopt a standardized definition of sepsis, and conduct multicenter prospective designs along with external validations.
-
Journal of critical care · Dec 2024
ReviewMachine learning for predicting mortality in adult critically ill patients with Sepsis: A systematic review.
Various Machine Learning (ML) models have been used to predict sepsis-associated mortality. We conducted a systematic review to evaluate the methodologies employed in studies to predict mortality among patients with sepsis. ⋯ ML models demonstrate a modest improvement in predicting sepsis-associated mortality. The certainty of these findings remains low due to the high risk of bias and significant heterogeneity. Studies should include comprehensive methodological details on calibration and hyperparameter selection, adopt a standardized definition of sepsis, and conduct multicenter prospective designs along with external validations.
-
Journal of critical care · Dec 2024
Meta AnalysisDoes extracorporeal cardiopulmonary resuscitation improve survival with favorable neurological outcome in out-of-hospital cardiac arrest? A systematic review and meta-analysis.
Extracorporeal cardiopulmonary resuscitation (E-CPR) may improve survival with favorable neurological outcome in patients with refractory out-of-hospital cardiac arrest (OHCA). Unfortunately, recent results from randomized controlled trials were inconclusive. We performed a meta-analysis to investigate the impact of E-CPR on neurological outcome compared to conventional cardiopulmonary resuscitation (C-CPR). ⋯ In this meta-analysis, E-CPR had a positive effect on survival with favorable neurological outcome and, to a smaller extent, on overall mortality in patients with refractory OHCA.
-
Journal of critical care · Dec 2024
ReviewCritical care management of hantavirus cardiopulmonary syndrome. A narrative review.
Hantaviruses, members of the Bunyaviridae family, can cause two patterns of disease in humans, hantavirus hemorrhagic fever with renal syndrome (HFRS) and cardiopulmonary syndrome (HCPS), being the latter hegemonic on the American continent. Andesvirus is one of the strains that can cause HCPS and is endemic in Chile. Its transmission occurs through direct or indirect contact with infected rodents' urine, saliva, or feces and inhalation of aerosol particles containing the virus. ⋯ The hemodynamic pattern of HCPS is not identical to that of cardiogenic or septic shock, being characterized by hypovolemia, systolic dysfunction, and pulmonary edema secondary to increased permeability. Given the lack of specific effective therapies to treat this viral infection, the focus of treatment lies in the timely provision of intensive care, specifically hemodynamic and respiratory support, which often requires veno-arterial extracorporeal membrane oxygenation (VA-ECMO). This narrative review aims to provide insights into specific ICU management of HCPS based on the available evidence and gathered experience in Chile and South America including perspectives of pathophysiology, organ dysfunction kinetics, timely life support provision, safe patient transportation, and key challenges for the future.
-
Journal of critical care · Dec 2024
ReviewVenoarterial extracorporeal membrane oxygenation in high-risk pulmonary embolism: A narrative review.
Emergent reperfusion, most commonly with the administration of thrombolytic agents, is the recommended management approach for patients presenting with high-risk, or hemodynamically unstable pulmonary embolism. However, a subset of patients with a more catastrophic presentation, including refractory shock and impending or active cardiopulmonary arrest, may require immediate circulatory support. ⋯ Retrospective studies and registry data suggest favorable clinical outcomes with the use of VA-ECMO as an upfront stabilization strategy even among patients presenting with cardiopulmonary arrest. In this review, we discuss the physiologic rationale, evidence base, and an approach to ECMO deployment and subsequent management strategies among select patients with high-risk pulmonary embolism.