Daily Anesthesiology Research Analysis

With sepsis remaining a leading cause of mortality, early identification of patients with sepsis and those at high risk of death is a challenge of high socioeconomic importance. Given the potential of hyperspectral imaging (HSI) to monitor microcirculatory alterations, we propose a deep learning approach to automated sepsis diagnosis and mortality prediction using a single HSI cube acquired within seconds. In a prospective observational study, we collected HSI data from the palms and fingers of more than 480 intensive care unit patients. Neural networks applied to HSI measurements predicted sepsis and mortality with areas under the receiver operating characteristic curve (AUROCs) of 0.80 and 0.72, respectively. Performance improved substantially with additional clinical data, reaching AUROCs of 0.94 for sepsis and 0.83 for mortality. We conclude that deep learning-based HSI analysis enables rapid and noninvasive prediction of sepsis and mortality, with a potential clinical value for enhancing diagnosis and treatment.

RATIONALE: Automated closed-loop systems may improve the adaptation of mechanical ventilatory support to an individual's ventilatory needs. They may also facilitate systematic and early recognition of the patient's ability to breathe spontaneously and come off the ventilator. This is an update of a Cochrane review originally published in 2013 and last updated in 2014. OBJECTIVES: To evaluate the benefits and harms of automated weaning systems compared with non-automated weaning methods in critically ill, mechanically ventilated adults and children. SEARCH METHODS: We searched MEDLINE ALL, Embase Classic+Embase, the Cochrane Library (Wiley), CINAHL (EBSCO), the Web of Science Core Collection, and trial registries on 2 February 2024. We checked the reference lists of included studies and relevant systematic reviews for other potentially eligible studies. ELIGIBILITY CRITERIA: We included randomized controlled trials (RCTs) evaluating automated closed-loop ventilator applications versus non-automated weaning methods (including non-protocolized usual care and protocolized weaning) in people aged over four weeks who were receiving invasive mechanical ventilation in an intensive care unit (ICU). OUTCOMES: Our critical outcomes were duration of mechanical ventilation (from randomization to successful unassisted breathing or death), mortality, ICU length of stay, and hospital length of stay. Our important outcomes included other ventilation durations, adverse events related to mechanical ventilation, and health-related quality of life. RISK OF BIAS: Two review authors independently assessed risk of bias using the Cochrane risk of bias tool RoB 1. SYNTHESIS METHODS: Two review authors independently extracted study data. We synthesized results for each outcome using meta-analysis (random-effects modeling). Subgroup and sensitivity analyses were conducted according to pre-established criteria. We used GRADE to assess the certainty of evidence for each outcome. INCLUDED STUDIES: This update included 62 trials (59 in adults, 3 in children) with 5052 participants (4834 adults, 218 children). The trials evaluated 10 commercially available automated closed-loop systems and one non-commercial system. Forty trials were conducted in mixed or medical ICU populations, the remainder in surgical ICU populations. SYNTHESIS OF RESULTS: Automated closed-loop systems probably reduce the duration of mechanical ventilation compared with non-automated weaning methods (mean difference [MD] -0.28 log hours, 95% confidence interval [CI] -0.36 to -0.20; I AUTHORS' CONCLUSIONS: Based on moderate-certainty evidence from 62 trials including over 5000 critically ill people (mainly adults), we found that automated closed-loop systems probably reduce the duration of mechanical ventilation and the length of ICU and hospital stay compared with non-automated weaning methods. Automated systems probably have little to no effect on mortality but probably reduce the need for reintubation, non-invasive ventilation, prolonged ventilation, and tracheostomy. Given the moderate-certainty evidence of benefit and no evidence of harm, the adoption of automated closed-loop ventilation systems into adult critical care clinical practice warrants consideration. There is a need for further adequately powered multi-center trials in adults and children. Future trials should include health-related quality of life among their outcomes. FUNDING: This review received no funding. REGISTRATION: The original review was registered with the Cochrane Database of Systematic Reviews, registration number CD009235. The original protocol, published in

BACKGROUND: Methods for reliably predicting hypotension in patients during general anesthesia induction are currently lacking. Deep inhalation has been shown to enhance the variability of the internal jugular vein (IJV). In this study, we aim to investigate the relationship between internal jugular vein variability (IJVV) during deep inhalation and the extent of blood pressure decrease during propofol induction, as well as the potential of utilizing IJVV as a guide for pre-anesthesia fluid resuscitation. METHODS: Before general anesthesia induction, bedside ultrasonic measurement was performed to evaluate the maximum diameter (IJVmax-D) and minimum diameter (IJVmin-D) of the IJV and the maximum cross-sectional area (IJVmax-A) and minimum cross-sectional area (IJVmin-A), and then calculated the IJV diameter variability (IJVV-D) and IJV area variability (IJVV-A). A receiver operating characteristic (ROC) curve was used to determine the diagnostic value of IJVV-D and IJVV-A for predicting propofol induced hypotension (blood pressure decreased ≥ 20%) and calculate the cut-off value. The following prospective randomized controlled trial aimed to compare the incidence of anesthesia-induced hypotension between the IJVV-D or IJVV-A guided fluid administration (Group A) and the standard fluid administration group (Group B) in patients with the variability value > optimal cut-off value. The occurrence rate of hypotension during the propofol induction period was observed and compared between the two groups. RESULTS: A total of 60 patients were included in the final analysis. A significant strong correlation exists between IJVV-A and the degree of blood pressure decrease during deep inhalation (r = 0.858, p < 0.001). The AUC of IJVV-A was 0.900 (95% CI 0.821-0.979, p < 0.001) with a cut-off value of 23.42% (sensitivity: 81.5%, specificity: 84.8%). At the same time, a total of 87 patients with IJVV-A > 23.42% during deep inhalation were included in the data analysis. The incidence of hypotension in Group A was 26.8%, compared to 63.0% in Group B, revealing a statistically significant difference (P < 0.001). CONCLUSIONS: A significant relationship was observed between IJVV levels during deep inhalation and the blood pressure decline following propofol induction. Administering IJVV-A guided fluid infusion can significantly reduce propofol-induced hypotension by keeping the IJVV-A less than 23.42% during deep inspiration. TRIAL REGISTRATION: Successfully registered on Clinicaltrials.gov on November 1, 2023 (NCT06112769) and on August 1, 2024 (NCT06641505).